The Blue Dot Issue 9: Artificial Intelligence and the Future of Education

EDITORIAL COMMITTEE 

ISSUE 9, 2018 

ANANTHA K. DURAIAPPAH 

Director 

PUBLISHED BY UNESCO MGIEP 

United Nations Educational, Scientific and Cultural 

Organization | Mahatama Gandhi Institute of 

Education for Peace and Sustainable Development 

35 Ferozshah Road, ICSSR, Building, 1 st Floor 

New Delhi 110001, INDIA 

THE BLUE DOT features articles showcasing UNESCO 

MGIEP’s activities and areas of interest. The magazine’s 

overarching theme is the relationship between education, peace, 

sustainable development and global citizenship. THE BLUE 

DOT’s role is to engage with readers on these issues in a fun 

and interactive manner. The magazine is designed to address 

audiences across generations and walks of life, thereby taking the 

discourse on education for peace, sustainable development and 

global citizenship beyond academia, civil society organisations 

and governments, to the actual stakeholders. 

NANDINI CHATTERJEE SINGH 

Programme Specialist - Science of Learning 

THE BLUE DOT is published biannually. 

SUBSCRIPTION 

The Blue Dot is available free of charge. 

To receive all future issues of the THE BLUE DOT, 

subscribe to bluedot.mgiep@unesco.org 

MANAGING EDITOR 

Akriti Mehra, UNESCO MGIEP 

DESIGN 

Prasun Mazumdar Design | www.pmdindia.in 

AKRITI MEHRA 

Communications Specialist 

© UNESCO MGIEP 

Disclaimer: The views and opinions expressed in this magazine 

do not necessarily reflect the official policy or position of 

UNESCO MGIEP. 

The image used on the cover of this issue of The Blue Dot is 

purely representational and conceptual in nature.

“Look again at that dot. 

That’s here. That’s home. That’s us. 

On it, everyone you love, 

everyone you know, everyone you ever heard of, 

every human being who ever was, 

lived out their lives. 

The aggregate of our joy and suffering 

thousands of confident religions, 

ideologies, and economic doctrines, 

every hunter and forager, every hero and coward, 

every creator and destroyer of civilization, 

every king and peasant, every young couple in love, 

every mother and father, hopeful child, 

inventor and explorer, every teacher of morals, 

every corrupt politician, every superstar, 

every supreme leader, every saint 

and sinner in the history of our species lived thereon 

a mote of dust suspended in a sunbeam.” 

CARL SAGAN 

PALE BLUE DOT: A VISION OF THE HUMAN FUTURE IN SPACE 

director’s 

message 

A 

child somewhere 

in “poor” India is 

having trouble with 

introductory algebra. 

This is new content for the child and the 

bigger problem the child faces is that the 

teacher is hardly ever in school. There is 

some hope, however, as s/he has good 

access to the Internet and to the many 

online teaching tools on the subject. 

Despite this, the child cannot afford a 

personalised online tutor and continues 

to struggle to understand the basic 

concepts. That, I believe, might be a 

close description of the current education 

system in India and many parts of the 

world.’ 

Now visualise a system where the 

child is given access to an online global 

intelligent education platform. As part 

of this system, the child is paired with 

a personalised BOT right after s/he 

enters school. This BOT develops a keen 

understanding of the child’s attributes and 

learning preferences by evaluating data of 

the child’s ongoing learning experiences.’ 

The BOT has access to a finite set 

of possible interventions – arising from 

learner-centric data derived from the 

experiences of millions of children 

learning algebra worldwide – to help the 

child overcome learning problems. The 

BOT can parse all the learner-centric data 

to identify specific interventions suited to 

the particular child who is its responsibility 

– functioning as a virtual, personalised 

Artificial Intelligence 

for Education 

tutor. Personalised learning, the holy grail 

of education, is a definite reality in this 

hypothetical scenario. As I perceive it, 

making this a reality for children today is a 

distinct possibility. 

Is this just in the realms of fiction or 

a potential reality? The recent growing 

importance of Artificial Intelligence (AI) 

seems to point such a scenario as a 

reality within our reach. However, there 

has also been growing concern over the 

ethical and moral use of AI coupled 

with a fear of AI and making many jobs 

redundant. Although the use of AI in the 

education sector is at the nascent stage, 

these fears are reflected as well and in 

particular to the rising sentiment that AI 

will replace teachers. 

Doucher and Evers, in their seminal 

book, “Teaching in the Fourth Revolution: 

Standing at the Precipice”, argue that AI 

will no doubt have a major impact in the 

way learning and teaching will happen in 

the future. There is no looking back. 

But what that future will look like is for 

us here at the present to guide that journey. 

One thing is for sure: the traditional roles 

of learners and teachers will evolve and 

will need to be more symbiotic. 

But before we go into details of AI’s role 

in education, some understanding of its 

scope is necessary. AI experts categorise 

three main types of AI: Strong AI, Weak 

AI and Artificial General Intelligence. 

These are well explored and the myth 

behind AI is discussed in this issue of 

The Blue Dot. These systems have been 

elaborated upon in detail and the myths 

surrounding them have been dispelled by 

some of the world’s foremost AI experts. 

The second part of the AI and 

education relationship explores how 

AI can be used to improve learning 

and make it relevant for 21 st century 

challenges. Articles written on this topic 

will explore whether the new world that 

I had proposed in the beginning of my 

message is achievable and if yes, what 

needs to be done to achieve that holy grail 

of education. 

This issue was based on one of the five 

main themes of our recently concluded 

TECH 2018 conference in Vizag, India, 

titled ‘AI and Education’, for which 

some of the world’s leading experts were 

invited to elaborate on the theme. Equally 

important to have is the discussion on the 

privacy and ownership of information that 

will be generated by learners and teachers. 

The use of information is a contentious 

issue that needs to be addressed early as we 

embark on the exciting and path-breaking 

journey that I have termed the The 

Internet of Education (TIE). I hope that 

this journey will culminate in education 

and knowledge becoming accessible to 

and beneficial for all of humanity. A High 

Level Policy Forum on development of 

guidelines was initiated at the TECH 

2018, after which the Vizag Declaration 

was adopted at the closing ceremony of 

the conference, calling for development of 

Guidelines for Digital Learning. 

We here at UNESCO MGIEP have 

already begun this journey by partnering 

with like-minded organisations and have 

developed a prototype platform called 

CHI (Collective Human Intelligence). This 

platform allows educators and learners 

to develop curriculum, lesson plans and 

assessments in an interactive, immersive 

and experiential environment, which is 

supported by AI that is able to provide 

feedback to students and educators of 

progress and suggestions to improve 

learning. 

ANANTHA KUMAR DURAIAPPAH 

Director, UNESCO MGIEP 

DIRECTOR’S MESSAGE 

0 1

FOREWORD 

Artificial Intelligence & 

Education & UNESCO’s 

Way Forward 

Currently, there is no universally 

recognised institutional framework 

on the ethics of AI, particularly in 

relation to education. 

FOREWORD 

MME AUDREY AZOULAY 

DIRECTOR-GENERAL, UNESCO 

T 

he permeation of new technologies and Artificial 

Intelligence (AI) is changing how we teach, live, 

work and learn. Undoubtedly, AI has the potential 

to widely benefit us and resolve various societal 

issues. At the same time, it also presents certain challenges that 

need to be addressed. Yet, if we are able to prepare ourselves to 

tackle these challenges, we will be able to benefit widely from this 

collective intelligence and the ‘Fourth Industrial Revolution’. 

In the field of education, AI is expected to have a 

significant impact, transforming the way teachers work 

by offering new teaching aids. It could revolutionise the way 

in which students learn through personalised learning and through 

greater access to knowledge thus also potentially facilitating more 

inclusive education. 

am very pleased to share that our Category I Institute in 

the Asia-Pacific, the UNESCO Mahatma Gandhi Institute 

of Education for Peace and Sustainable Development 

(MGIEP), has taken the lead on SEL and exploring ways 

AI can be harnessed to integrate SEL into our Education 

systems. 

I am particularly pleased with the progress that has been made with 

the development of an open-source indigenously-designed and 

collaborative curriculum development and learning platform that 

allows educators and students to innovate, develop and implement 

curriculum. This Collective Human Intelligence platform will use 

advanced AI and data analytics tools to personalise the learning 

experience for each user, as well as address data ownership, privacy 

and security issues. 

Clearly, the possibilities are tremendous. However, there are some 

associated risks, particularly those related to individual privacy 

and ethics. UNESCO, by its mandate, is ideally placed to 

play a key role by supporting Member States to increase 

accessibility of technology, promoting the use of opensource 

AI platforms and tools that encourage learning. 

Additionally, UNESCO’s role will be crucial to rethink educational 

curricula, by bringing into focus social and emotional learning to 

build the emotional intelligence of future generations of learners. 

Currently, there is no universally recognised institutional framework 

on the ethics of AI, particularly in relation to education. It is 

UNESCO’s duty, with the support of Member States, 

to lead a global debate in order to establish a globally 

accepted framework of ethical principles and we have 

already begun work in this area. 

Furthermore, I welcome the collaboration between the Institute 

and the State Government of Andhra Pradesh, India, as well 

as other partners and experts in the field of EdTech to establish 

global guidelines for the exponentially growing field of digital 

learning tools and methodologies. These guidelines will help the 

development of digital pedagogical tools to effectively harness this 

new technology for the betterment of society. I also look forward 

to the outcomes of the Institute’s annual international EdTech 

conference, TECH 2018, for which AI and Education was a 

key focus area. I wish the Institute all the best in its endeavours 

to mainstream SEL through the use of AI: a vital issue as we 

deliberate on the ethics of AI. 

The ninth issue of The Blue Dot, focuses on AI and the future of 

education. It explores debates on the use of AI in education and the 

potential it offers to complement existing education systems and 

teachers, as well as in facilitating personalised learning. Indeed, I 

ISSUE • 9 

0 3

FOREWORD 

Approach to 

Artificial Intelligence for 

Education and Training 

AI is concerned with understanding 

intelligence, by investigating the 

design of machines that have 

intelligent abilities. 

FOREWORD 

A 

rtificial Intelligence (AI) is no 

longer just a Hollywood invention, 

it is already impacting our lives. 

From truck drivers, whose role may be 

replaced by autonomous vehicles, to paralegal workers who can 

be outsmarted by AI software, the world of work is changing with 

increasing pace. The implications for education and training are 

significant and far reaching. We need to prepare people for a future 

that will likely be evolving quickly, and not everyone will have a 

positive experience. However, it is not all doom and gloom. We 

can also reap huge benefits from AI if we approach it 

intelligently. 

to software that can support students with individual tutoring for 

their school and university studies, and learning platforms that 

provide educational support informed by neuroscience. We can 

track students’ interactions, from every mouse movement and each 

keystroke and use AI to analyse each student’s performance in 

detail. This analysis provides personalised learning for the student 

and detailed feedback about the learning status of every student in 

their class for teachers. Well-designed AI can help us tackle some 

of our toughest educational challenges, such as global teacher 

shortages and the achievement gap between the more and less 

advantaged members of the population. 

But what exactly is AI, and how can we approach it intelligently? 

ROSE LUCKIN 

PROFESSOR,UCL Knowledge Lab 

AI is concerned with understanding intelligence by 

investigating the design of machines that have intelligent 

abilities. AI has gained traction recently because of the ‘perfect 

storm’ brought about by the availability of large quantities of data, 

powerful computer processing with large clouds of cheap, accessible 

storage and AI algorithms that can learn. These 3 factors: data, 

computing (power and storage), and sophisticated AI means that we 

can build vast AI ‘knowledge’ bases that can be sifted and searched 

with stunning accuracy to pinpoint the minutiae of detail required 

to diagnose a disease, process an image, beat world champion game 

players and to help us navigate through our complex physical and 

virtual worlds. 

The intelligent way to approach AI is to focus on planning 

educational and training systems that use AI wisely to 

prepare people to work and live with AI. The situation is 

summarised in Figure 1 

This figure illustrates that we need to: 

1. Engage educators and trainers in working with AI developers to 

build AI technology that is specifically designed for education and 

training. This AI development has already started: from experiential 

digital learning driven by AI powered virtual role play simulations, 

2. Educate people about AI, so that they can use it successfully, 

ethically and knowledgeably. For some people, this understanding 

about AI needs to be more extensive, to train people to: make 

responsible ethical decisions about how and when AI should be 

used; or develop the next generation of AI technology. Educating 

people about AI, can also be assisted by well-designed AI systems. 

3. Prioritise the development of our own human intelligence so that 

we focus attention on the elements of intelligence that are not (yet) 

automatable with AI, such as social intelligence and metacognitive 

intelligence. This too can be assisted by the smart use of welldesigned 

AI to track and support our learning and increase our 

intelligence. For more detail about this, please see my 2018 book: 

Machine Learning and Human Intelligence. 

The thoughtful design of AI approaches to education and 

training has the potential to provide significant benefits 

to educators, learners, parents and managers. It is a reason 

for great excitement and concerted intelligent effort. 

ISSUE • 9 

0 5

CONTENTS 

ISSUE 

36 

44 

9, 2018 

Director’s Message 

Artificial Intelligence for Education 

By Anantha Kumar Duraiappah | Director, UNESCO MGIEP 

Foreword 

• Artificial Intelligence & Education & UNESCO’s 

Way Forward 

By Audrey Azoulay | Director-General, Unesco 

01 

02 

• Approach to Artificial Intelligence for Education & 

Training 

By Rose Luckin | Professor, UCL Knowledge Lab 

• Declaration of Independence II 

By Dan Shefet 

• Dreams & Reality: How AI will change Education 

By Harri Ketamo 

30 

FEATURE ARTICLE 

• CHI – a Knowledge Sharing Digital Platform Powered 

by Machine Learning 

Saurabh Roy | Chief Technology Officer, UNESCO MGIEP 

INTERVIEWS 

40 

• Interview with Dr. Avik Sarkar, Head – Data Analytics 

Cell at NITI Aayog, Govt. of India 

YOUTH VOICES 

• Can Artificial Intelligence Help Us Achieve Universal 

Quality Education? 

By James Hodson 

• Artificial Intelligence for Education 

By Soma S Dhavala 

50 

06 

42 

ACTIVITY BULLETIN 

OPINIONS 

What we have been up to at UNESCO MGIEP 

• The Digital Transformation of K-12: A Viewpoint 

By Cathie Norris & Elliot Soloway 

• Peace Machine as a Re-emerging Gandhi 

By Timo Honkela 

• Promoting Peace: How Can Data Science Assist? 

By Srujana Merugu 

COVER STORY 

• Personalising ‘Learning’ - Can AI Promise 

Customised Education For ‘Humanity’ 

Nandini Chatterjee Singh & Raunak Jain 

TECH 2018 

What to Expect at TECH 2018? 

COMIC STRIP 

Learning with AI 

60

opinion 

Expert 

Perspectives 

The Digital Transformation 

of K-12: A Viewpoint 

OPINION 

Artificial Intelligence and the Future of 

Education 

Artificial Intelligence (AI) has caught the 

imagination of the world and has been gaining 

popularity in the business sector. To date, the 

use of AI in education has been limited and at 

times contested; most arguments against it centre 

on how it will make teachers redundant and 

learning more automated than it is currently. 

Rather than replacing teachers and making 

learning impersonal, AI could take learning to a 

completely new level. Read what academicians, 

policymakers, practitioners and researchers have 

to say about the use of Artificial Intelligence 

(AI) in Education 

CATHIE NORRIS, Regents Professor, Department of Learning Technologies, College of Information at 

The University of North Texas 

ELLIOT SOLOWAY, Arthur F. Thurnau Professor, Department of Cse, College Of Engineering, at 

The University Of Michigan 

Dr. Cathleen Norris is the Regents Professor & Chairperson, Department of 

Learning Technologies, College of Information, University of North Texas. Dr. 

Norris’ 14 years in K-12 classrooms – and receiving Dallas’ Golden Apple Award 

– has shaped her university R&D agenda: developing resources to support 

K-12 teachers as they move into 21 st century classrooms. From 1995-2001, 

Dr. Norris was President of the National Educational Computing Association 

(NECA), and led its merger with ISTE, the International Society for Technology 

in Education, creating the largest, international organisation for technologyminded 

educators in the world. Dr. Norris was Co-President/President of ISTE 

from 2001-2004. 

Dr. Elliot Soloway is an Arthur F. Thurnau Professor, Department of Computer 

Science and Engineering, College of Engineering, with appointments also in 

the School of Education and School of Information, University of Michigan. For 

15+ years, Dr. Soloway’s R&D efforts have been guided by the vision that mobile, 

low-cost, networked devices are the only way to truly achieve universal 1:1 in 

schools – all across the globe. In 2001, the UMich undergraduates selected him 

to receive the “Golden Apple Award” as the Outstanding Teacher of the Year at 

the University of Michigan. In 2004 and in 2011, College of Engineering HKN 

Honor Society students selected Dr. Soloway to receive the “Distinguished 

Teacher of the Year Award.” In 2001, Norris and Soloway founded GoKnow, 

Inc., one of the first mobile learning companies; over 40,000 students 

worldwide used GoKnow’s Mobile Learning Environment. In 2012, Norris and 

Soloway founded the Intergalactic Mobile Learning Center, and with support 

from private (e.g., Qualcomm, Google, George Lucas Education Foundation) 

and public (e.g., National Science Foundation) sources, has published the 

Collabrify Suite of Apps and the Collabrify Roadmap System – free, deviceindependent, 

tools, used in classrooms all around the world. 

0 7

OPINION 

The Digital Transformation of K-12: A 

Viewpoint 

• “Most 1 school leaders know we’re in the midst of a digital 

transformation, with a size and scope larger and more significant 

than anything we’ve seen before…. Educators know they have to 

change but are challenged by the convergence of technology and 

pedagogy. What technologies can affect learning and experience? 

How can we get all educators on board? How can we build a case 

to support our goals? What’s our future vision?” 

The above quote sums up the status of the “digital 

transformation”, or DX 2 to the cognoscenti, in K-12: desire – at 

least some – is there, but action – for the most part – is lacking. 

Arguably, the digital transformation of K-12 education is the most 

important task ahead for us in education. It is towards developing 

a deeper understanding of what digital transformation is, why it 

needs to be undertaken, and how it can be accomplished, that we 

write this Blue Dot article. So, let’s begin with the “what.” 

While those quotes had “business” not 

“education” in mind, the reality is this: 

educating students is as much a business 

as banking, retailing, entertaining, etc. 

are! 

century skills – creativity, communication, collaboration, and 

critical thinking. And, while “driving revenue” is not a focus of 

public education, the public purse snaps shut if value is not being 

delivered! 

• “… using digital technologies” — The new acronym for the 

range of technologies encompassed by the term “digital” is 

SMAC 5 : social, mobile, analytics and cloud technologies. But 

let’s make it “SMAAC” – where the second A represents “AI” or 

artificial intelligence. 

OPINION 

What is Digital transformation? 

• “Digital transformation 3 is a fundamental shift in how a firm 

delivers value and drives revenue.” 

• “Digital transformation 4 involves using digital technologies to 

retool business models and processes to make them more efficient 

or effective.” 

While those quotes had “business” not “education” in 

mind, the reality is this: educating students is as much a 

business as banking, retailing, entertaining, etc. are! 

Let’s unpack those dense sentences: 

• “… a fundamental shift …” Not just a “shift” but a “fundamental 

shift.” Not digital transition but digital transformation. Words do 

count here and what is being proposed is truly a dramatic and 

“fundamental” break with the past. Analog, paper-and-pencil 

technologies, while comfortable and effective – for some – are not 

the building blocks of the future. Digital is based on bits; analog 

is based on atoms. Atoms and bits are different: duplicating atoms 

is expensive; duplicating bits is trivial; changing atoms – erasing a 

word in a sentence – is messy and time consuming; changing bits – 

plopping in a new word into a sentence – is easy, quick, and cheap. 

Indeed, we are only beginning to appreciate the implications of a 

world based on bits, not atoms. 

• “… how a firm delivers value and drives revenue.” Delivering 

“value” is absolutely the top priority in schools. While that value 

shares much with the past – education must develop students to be 

ethical individuals, 21 st century students must develop 21 st 

• “… retool business models and processes” — Simply put 

a “business model” is the way a company makes money. For 

example, one of Apple’s core business models is manufacturing 

and selling iPhones. But, if the rumors 6 are true, Apple is creating 

a new streaming service that includes original TV content, so that 

collecting subscription fees is an example of Apple retooling its 

business model and processes. In education, rather than a teacher 

standing in front of the class and delivering a lecture, while 

students copy down on paper the teacher’s words, schools will need 

to develop “digital pedagogies” that exploit SMAAC and provide 

students with the personalised learning experiences that they are 

demanding. 

• “… make [the business models and processes] more efficient or 

effective” —The terms “efficiency” and “effective” are at the core 

of digital transformation. We will explore those terms when we 

discuss the “how” of digital transformation. 

Why Is Digital Transformation Needed? 

• “Nine 7 out of 10 businesses today are undergoing some form 

of digital transformation and customer experience is a stated top 

priority.” 

While there is debate 8 that the 1983 “Nation At Risk” report 

may have portrayed American education in an overly poor light, 

there is no question that there is considerable room for improving 9 

student achievement in U.S. schools. Indeed, in education today, 

“customer experience is the priority!” The following title 10 of an 

article paints a clear picture: 

• “2018 Digital School Districts Survey: Respondents Use 

Technology to Put Students First” 

The survey 11 , tapping into schools in “small, mid-sized and 

large districts in 24 states” conducted by the Center for Digital 

Education found 12 that: 

More than nine in 10 respondents to a 

survey on the topic reported that in the 

pursuit of encouraging personalised 

learning, their districts 1) provide software 

or digital curriculum to classrooms … 2) 

provide computing devices to classrooms 

• “The top learning priority in education for technology use is 

personalised learning. 

And the only way to provide personalised learning is via 

technology 13 : 

• “More than nine in 10 respondents to a survey on 

the topic reported that in the pursuit of encouraging 

personalised learning, their districts 1) provide software 

or digital curriculum to classrooms … 2) provide 

computing devices to classrooms…” 

As we will describe in the “How” section below, there are several 

ways to provide personalised learning – but they all employ 

technology. 

Besides needing to improve “customer experience,” digital 

transformation is needed because those customers – the students 

of today and most assuredly of tomorrow – are expecting it! 

Today’s students are virtually born with a “phone” – the students 

of today do not call their mobile devices “smartphones” but simply 

“phones” – in their hands. Phones: The penetration of personally 

owning a smartphone is now down 14 to the 12-13 year olds in 

5 th grade 15 . Yes, today, the research 16 is mixed on whether using 

ISSUE • 9 

0 9

OPINION 

computers for learning leads to increased student achievement. 

But, just as entertainment, medicine, retailing, farming, banking, 

etc., etc., are not going back to analog technologies, education is 

not going back to paper-and-pencil technologies. Educators will 

learn how to use digital devices to enhance learning; it’s only a 

matter of time. 

How is Digital Transformation Realized? 

In carrying out DX, the consensus is that technology itself is 

important: 

• “Digital transformation is often driven by the IT organization. … 

The technology is actually the center of what growth is built on. 

The technology implications are driving the future of the business 

itself.” 

And the core technology driver will be AI – Artificial Intelligence: 

• “If [your company doesn’t 17 ] have an AI strategy, you’re going to 

die in the world that’s coming. … more and more AI interfaces will 

provide natural language inputs and outputs, and proactively make 

decisions on the user’s behalf … “ 

AI – or, more accurately 18 , ML – machine learning 19 – is the latest, 

cure-all, technological hype in a long line of technological hype – 

some successful (2G, 3G, 4G), some not (have you seen a 3D TV 

lately?) AI is not a new technology; researchers have been 

exploring AI since the early days 20 of computing. But, 

thanks to the exponential growth of the chip industry, as predicted 

so presciently by Gordon Moore in 1965, AI has come on the 

scene now only because computers are finally 21 powerful enough – 

at everyday prices – to actually run the computation necessary to 

implement “AI.” 

As we have argued previously 22 : 

• “Machine learning techniques underlying personalised 

learning will disrupt, in the Christensen-sense, your 

child’s school in the near term. You can take that 

prediction to the bank!” 

For example, in early math education, in grades 1-3, companies 

such as Dreambox 23 and Knewton 24 are providing personalised 

instruction 25 systems that do result in improved test scores. That 

said, we urge the interested reader to review a “confession 26 ” 

by Larry Berger, CEO of Amplify 27 , Inc., who has been trying 

to build personalised learning systems for over 10 years. Berger 

argues that there are a number of issues that prevent the 

development of a personalised learning system for all grades and 

all subjects and for all students. 

But, let’s not throw the baby out with the bath water: machine 

Machine learning techniques underlying 

personalised learning will disrupt, in the 

Christensen-sense, your child’s school in 

the near term. You can take that prediction 

to the bank! 

learning algorithms can provide useful learning analytics to 

teachers so they can personalize the learning for their students. 

While learning analytics can be deceiving at times, 

• ”… it has become clear to many that the use of data is critical 

to how ‘educators 28 evaluate their practices and monitor students’ 

academic progress.’” 

As well, ML can provide more “natural” ways to interact with 

computers. For example, the explosion of voice-based interfaces 

(e.g., Apple’s Siri, Amazon’s Alexa 29 ) point to a future where 

voice can be used for both fast and sustained interaction. Integral 

to voice-based interfaces is natural language understanding 30 

by computer. Yes and no. As an experiment, we went into a 

kindergarten (children of age 4, 5, 6) and tried to use voice and 

natural language software to analyse these children’s utterances. 

While the software failed almost every time – kindergartners do 

not speak in grammatical sentences, do not enunciate clearly, and 

make up words willy-nilly – we were assured by ML engineers that 

if they had 5,000 hours of recorded speech from these youngsters, 

they could build an effective voice-based, natural language system 

for those youngsters! Perhaps! 

ML can also support the production of curriculum that serves as 

the heartbeat of a classroom. 

• Chi.Buzz – Developed at the UNESCO Mahatma Gandhi 

Institute of Education for Peace and Sustainable Development 

(MGIEP), CHI – Collective Human Intelligence – is a platform 

for creating digital courses that employs ML to support the 

development and enactment of digital courses. 

• Collabrify Roadmap – Developed at the Intergalactic Mobile 

Learning Center (IMLC), at Roadmap.center educators can create 

digital curriculum, Roadmaps, visual, concept-map-like structures, 

that guide students through a lesson. Finding content on the 

Internet (in the form of OER – Open Education Resources) for 

the learning activities in those Roadmaps is a challenging, timeconsuming 

activity. We are exploring how to incorporate ML to 

provide curriculum developers with suggestions about relevant 

OER content. 

Games, that combine novel, graphical interfaces with ML 

algorithms in the context of provocative story-lines, may ultimately 

lead the way in the digital transformation of K-12. Presenting 

learning opportunities in the context of solo and collaborative 

game play, may well better align with the motivating factors for 

today’s learners. As the learners of today play games outside of 

school, it makes good sense to bring the mode of learning inside 

of school, as well. 

We truly believe that a bet against technological change is a really 

bad bet; new digital technologies will be invented and they will 

enable dramatic transformations. But, while ML may have finally 

conquered 31 the game of Go and bested our best human player 32 , 

delivering personalised instruction 33 , diagnosing disease, predicting 

financial – and weather – changes, etc. are highly challenging 

and will require our absolutely best thinking and absolutely best 

technology. These are early days for ML; we urge patience and 

caution 34 . 

Now, digital transformation isn’t just about technology! In all the 

articles about DX, success in DX involves people, organizational 

culture, and leadership: 

• “Hire 35 the right people to bring about change. Create a culture 

that makes data-based decisions.” 

• “… success 36 requires CEOs to develop the right leadership 

capabilities, workforce skills and corporate cultures to support 

digital transformation….” 

• Easy to say! In today’s schools, teachers are predominately 

drawn from the “baby-boomer” and “Gen X” generations, who 

are digital immigrants, while the students in their classes are 

millennials who are digital natives. 

And 37 , 

• “ … research shows that student teachers will initially teach the 

way they were taught.” 

While teachers can learn and change 38 , a recent study 39 found that: 

• “A majority (70-78%) of teachers expressed discomfort with 

newer teaching practices that rely on technology” 

Daunting! With all the professional development that has 

been showered upon today’s teachers, “a majority” are still 

uncomfortable using technology to facilitate collaboration, enable 

project-based learning, personalize learning, etc. 

Concluding Remarks 

Digital transformation is not a one-time event; rather it is as 

ongoing process: 

• “Digital 40 is a journey that requires [organisations] to be agile” 

And, 

• “Digitally 41 maturing [organisations] cultivate a culture that 

supports change and attracts talent. Commitment and leadership 

are required. To succeed, you have to be dedicated to investment, 

change, talent development, and have a clear vision… and be 

committed to it.” 

In the business world: 

• “… all companies 42 must transform to be digital companies to 

remain competitive.” 

In K-12? The economic forces driving businesses are not 

necessarily the same ones that drive K-12 education. What 

schools, what districts will invest the time and resources to digitally 

transform? Interesting questions – stay tuned! 

OPINION 

ISSUE • 9 

1 1

OPINION 

REFERENCES 

REFERENCES 

OPINION 

1 

https://www.cisco.com/c/dam/en_us/solutions/industries/docs/education/digital-learningenvironment.pdf 

2 

https://www.avaya.com/en/faqs/faq-detail-digital-transformation-all-about/ 

3 

http://dialog.vonage.com/industry-trends/the-5-key-drivers-of-digital- transformation-today 

4 

https://www.techrepublic.com/article/digital-transformation-technology-and- servicesspending-is-on-the-rise/ 

5 

https://searchcio.techtarget.com/definition/SMAC-social-mobile-analytics-and-cloud 

6 

https://www.macrumors.com/2018/06/27/apple-streaming-service-bundle-tv-music-news/ 

7 

https://www.avaya.com/en/faqs/faq-detail-digital-transformation-all-about/ 

8 

https://www.npr.org/sections/ed/2018/04/29/604986823/what-a-nation-at-risk-got-wrongand-right-about-u-s-schools 

9 

https://www.npr.org/sections/ed/2018/04/29/604986823/what-a-nation-at-risk-got-wrongand-right-about-u-s-schools 

10 

http://www.govtech.com/education/k-12/2018-digital-school-district-survey-respondentsuse-technology-to-put-students-first.html 

11 

http://www.govtech.com/education/k-12/2018-digital-school-district-survey-respondentsuse-technology-to-put-students-first.html 

12 

https://thejournal.com/articles/2018/08/13/personalized-learning-top-priority-for-techusage-in-k12.aspx 

13 

https://thejournal.com/articles/2018/08/13/personalized-learning-top-priority-for-techusage-in-k12.aspx 

14 

https://www.cnn.com/2017/12/11/health/cell-phones-for-kids-parenting-without-bordersexplainer-intl/index.html 

15 

https://techcrunch.com/2016/05/19/the-average-age-for-a-child-getting-their-firstsmartphone-is-now-10-3-years/ 

16 

https://thejournal.com/articles/2015/09/21/oecd-report.aspx 

17 

http://dialog.vonage.com/industry-trends/the-5-key-drivers-of-digital-transformation-today 

18 

http://thelearningcounsel.com/article/ai-vs-machine-learning-and 

education%E2%80%99s-future 

19 

https://medium.com/datadriveninvestor/differences-between-ai-and-machine-learning-andwhy-it-matters-1255b182fc6 

20 

http://sitn.hms.harvard.edu/flash/2017/history-artificial-intelligence/ 

21 

https://thejournal.com/articles/2017/09/18/ai-is-on-fire.aspx 

22 

https://thejournal.com/Articles/2017/07/24/A-Christensen-style-Disruption.aspx?Page=1 

23 

http://www.dreambox.com/white-papers/rti-toolkit?ppc&&gclid=Cj0KCQiA3b3gB 

RDAARIsAL6D-N-xZaXNmroovmtNBtu6x_ZRazIbLkJ2qaySOzEt8p1wEXIT-tEq- 

BkaAkR7EALw_wcB&gclsrc=aw.ds 

24 

https://www.knewton.com/the-power-of-altas-adaptive-technology/ 

25 

https://thejournal.com/articles/2014/12/04/personalized-learning.aspx 

26 

http://blogs.edweek.org/edweek/rick_hess_straight_up/2018/02/a_confession_and_a_ 

question_on_personalized_learning.html 

27 

https://www.amplify.com/ 

28 

https://ies.ed.gov/ncee/wwc/Docs/PracticeGuide/dddm_pg_092909.pdf 

29 

https://developer.amazon.com/alexa-skills-kit/vui 

30 

https://www.nytimes.com/2018/11/18/technology/artificial-intelligence-language.html 

31 

https://spectrum.ieee.org/semiconductors/processors/the-multiple-lives-of-moores-law 

32 

https://techcrunch.com/2017/05/23/googles-alphago-ai-beats-the-worlds-best-human-goplayer/ 

33 

https://nepc.colorado.edu/publication/newsletter-personalized-learning-121118 

34 

http://www.rogerschank.com/fraudulent-claims-made-by-IBM-about-Watson-and-AI 

35 

https://hbr.org/2016/03/the-industries-that-are-being-disrupted-the-most-by-digital 

36 

https://www.weforum.org/agenda/2016/01/digital-disruption-has-only-just-begun/ 

37 

https://web.uvic.ca/~thopper/Cupr/Archived/waytaught.htm 

38 

https://gianfrancoconti.wordpress.com/2015/12/23/why-teachers-teach-the-way-they-dopart-1/ 

39 

https://thejournal.com/articles/2018/07/30/educators-largely-uncomfortable-with-newertech-based-teaching-practices.aspx 

40 

https://www.zdnet.com/article/digital-transformation-retooling-business-for-a-new-age/ 

41 

https://sloanreview.mit.edu/projects/achieving-digital-maturity/ 

42 

http://dialog.vonage.com/cloud/5-myths-about-digital-transformation 

ISSUE • 9 

1 3

opinion 

Peace Machine as a 

OPINION 

Re-emerging Gandhi 

TIMO HONKELA 

A 

rtificial Intelligence (AI) has been under 

development for more than 50 years. 

Many ideas central to AI have been under 

consideration from very early on. The basic 

idea is to model or mimic human intelligence, including decision 

making, problem solving, language and speech understanding, 

machine learning, pattern recognition and a number of other 

characteristics. Even emotions and creativity have been taken 

into consideration. A common idea among AI researchers is to 

see the developments as models of the true phenomenon, not as 

independent achievements. Computers are not conscious and 

emotional in a human way but successful analysis and modelling 

has causal effects in the world of living humans. 

Timo Honkela serves as a Professor at the University 

of Helsinki, Department of Digital Humanities. 

Earlier positions include Professorships at the 

University of Art and Design Helsinki and Helsinki 

University of Technology, where he received his PhD 

in 1998. Dr. Honkela has served as Chair of Finnish 

Artificial Intelligence Society and as Executive 

Board Member of the European Neural Networks 

Society. His interests are in philosophical topics 

regarding human existence, emergence in relation 

to conceptual systems, and cultural evolution. 

The societal impact of AI has grown considerably during 

recent years. There are multiple reasons for this development. 

The use of artificial neural networks and statistical machine 

learning are used to analyse large data collections. These collections 

are like artificial experiences: sounds, spoken and written language, 

images and videos. The systems have growing cognitive capacity, 

ability to model human intellect based on the availability of 

increased amount of data and computing power and data storage. 

Earlier the systems were programmed with the aim to transfer 

expertise into them in the form of rules. Current solutions are 

mainly based on programming systems to learn. 

The AI developments are mostly directed to industrial and 

commercial purposes. The steps taken in this may lead into more 

efficient provision of products and services. An emerging challenge 

is that many professionals lose their work. This leads into problems 

in the societies as finding the work place as a replacement is not 

straightforward. The AI methodologies are also used to develop 

increasingly efficient weapons. Overall, many people feel that the 

The Peace Machine is a concept of 

applying AI with a range of technologies 

and disciplines understanding to promote 

peaceful conditions in the world. 

world is becoming increasingly frightening, 

also due to growing environmental 

concerns. 

The Peace Machine is a concept 

of applying AI with a range of 

technologies and disciplines 

understanding to promote peaceful 

conditions in the world. The 

main areas are (1) increasing mutual 

understanding among people, (2) 

promoting awareness of one’s own and 

others’ emotions, and (3) developing fair 

conditions on a large scale. The Peace 

Machine is not one device or system 

but a wide range of modules, tools 

and applications. They are mostly 

intended to be available for and used by 

large numbers of people. 

Development of the concept of the 

Peace Machine began in early 2017. The 

author’s own personal history adds another 

dimension to this. This includes 30 years 

of research on AI and being diagnosed 

with the worst kind of brain cancer. The 

tumor was operated upon in December 

2014 but left the author blind in the right 

eye. Nevertheless, the fact that life gave the 

author a reprieve led to the idea of a book 

on the Peace Machine. Many interviews 

to newspapers, radio and television 

resulted in an outpouring of support, and 

thousands of people helped the author 

write the book “The Peace Machine - 

The Testament of an AI Researcher”, 

which was published in October 2017 in 

the Finnish language. The book is now 

being translated to English and should be 

published by 2019. Another effect of this 

was that a society in Helsinki came into 

being in September 2018 to coordinate 

research, development, and international 

collaboration on this topic. 

ISSUE • 9 

1 5

OPINION 

Increasing mutual understanding among 

people 

In general, it is hard to arrive at mutual understanding. 

It is particularly difficult to achieve this when people have 

different backgrounds, education, experience, cultural 

contexts, etc. People not only have different values and 

opinions, the same words also hold different meanings 

for them.This is quite obvious for those who work in multicultural 

or international organisations. Information technology has not 

been aligned to deal with this phenomenon as computers are 

programmed with clear definitions of meaning. The development 

of AI and information technology has to be guided by careful, 

philosophical understanding. The systems that conduct meaning 

negotiations take into account the contexts in which people express 

Luckily, in addition to machine 

translation, tools for meaning negotiation 

can be developed to help in managing 

situations in which differences in 

experience, education and values have 

caused the conceptual systems of people to 

be different. 

themselves. The meaning of a word is not carved in stone but it is 

dependent on the linguistic, external and even internal context. Us 

humans can deal with this to some extent but information systems 

usually cannot deal with this at all. The idea of unambiguous 

meaning and crisp interpretation has been a guiding principle for 

many strands of philosophy and for computer science. Regrettably, 

it is not practical because we humans form heterogeneous 

groups.. Luckily, in addition to machine translation, tools 

for meaning negotiation can be developed to help in 

managing situations in which differences in experience, 

education and values have caused the conceptual 

systems of people to be different. The ways in which people 

understand language cannot be solved by education because us 

human cannot be reprogrammed like machines. The machines 

can, however, serve as our interpreters if we so wish. 

Good quality machine translation hypothetically improves 

peaceful conditions in the world as people can encounter over 

language borders. On the other hand, meaning negotiations may 

be needed in addition to straightforward machine translation. 

Google has succeeded in improving the quality of its machine 

translation system by taking into use neural network methodology, 

in particular deep learning systems with multiple “intuitive” 

information processing layers. The Peace Machine is intended 

to be available and useful for people all around the world. To 

reach necessary levels of trust, the systems must use open source 

software. A system that can have substantial effect on people 

cannot be closed within the digital walls of the system. The 

citizens and the governments, or their trusted representatives need 

to be able to see what is within the system, whether they live in 

India, USA, China, Russia, Iran, Israel, Iceland or anywhere else. 

Promoting awareness of one’s own and 

others’ emotions 

An important part of human decision-making is based 

on our emotions. Emotions help us to make quick decisions 

when swift action is required. On the other hand, our emotions can 

become tuned into some less optimal states or reaction tendencies. 

From the point of view of promoting peaceful conditions, 

orientation towards fear and anger is problematic. They prepare 

us to be aggressive and violent. They can be used to manipulate 

us to be available for violent actions even when these actions are 

against our best interests. It can be also asked if violence is ever 

truly well grounded. The true problems in the world are related 

to poverty, health and the environment. Violence and war may 

sometimes be seen as a means to break through some situation but 

there are other solutions. Machines do not have emotions 

but they can analyse our emotional states and related 

individual and social processes. We can develop our wisdom 

with the AI systems that would make us more aware of underlying 

mental processes that may be problematic, for instance, due to 

childhood trauma or culturally embedded reaction patterns based 

on cultural patterns established over generations. It may sound 

paradoxical but we could use computers to help us to achieve 

progress in these matters. A high level of transparency is needed 

Machines do not have emotions but they 

can analyse our emotional states and 

related individual and social processes. 

in the development of these kinds of systems. Machine learning 

is not a straightforward solution in all of this because machine 

learning, in its most common forms, is based on data that may 

also lead to problematic solutions, not only good ones. Therefore, 

intentionality is an essential concept: what do we wish to achieve, 

why and how. This is only to a limited way a technological 

question but much more a societal and ethical question. 

AI can be used and further developed into 

directions that support reaching goals that 

Mahatma Gandhi would be happy and 

proud of. 

Developing fair conditions on a large scale 

What is fair can be seen as a question of meaning negotiation. 

Moreover, it is an essential question when we consider human 

condition and our societies. Situations in different parts of the world 

are history dependant to a high degree. Not all people consider 

all people to be free and have basic human rights regardless of 

their gender, race and position. From the systems theoretical 

point of view, equality is a problematic notion. One possibility 

of approaching this is to consider distributions and contextual 

representations of reality. Like emerging new medicine provides a 

chance to take care of each person individually, the societies can 

be made better for all taking each person individually, providing 

improved conditions for all. This can help in the promotion of 

freedom, associated with increasing level of responsibility. The 

developments do not need to be directed by a small number 

of people like it is done nowadays. Namely, natural language 

processing and machine learning is making it possible to organise 

true meetings among very large number of people. A meeting of a 

million or billion people is not a voting arena but provides chances 

for true discussions when matters are considered among people 

at a grass root level. People will be able to craft solutions with 

unforeseen capacity and level of participation. 

Some of the goals for development may sound utopian. However, 

the content presented in the book does not relate to predictions but 

to goals and plans. AI can be used and further developed 

into directions that support reaching goals that 

Mahatma Gandhi would be happy and proud of. 

OPINION 

ISSUE • 9 

1 7

opinion 

Promoting Peace: How Can 

OPINION 

Data Science Assist? 

SRUJANA MERUGU 

Srujana is a Machine Learning researcher and 

consultant with over a decade of experience. She 

currently works as a founding member of a stealth 

mode startup. Prior to that, she was a principal data 

scientist at Flipkart (Bangalore) and a volunteer 

for Ekstep, an education startup. She has been 

employed with Machine Learning groups at Amazon 

(Bangalore), IBM Research (Bangalore/New Delhi), 

and Yahoo Research (Santa Clara). Srujana has 

published her work in several conferences and 

journals on data mining/machine learning and is 

the recipient of multiple best paper awards. She 

received her M.S. and Ph.D. from the University 

of Texas at Austin and her B. Tech. degree from IIT 

Madras. 

P 

eace has long been a professed goal of the 

mankind, but it has not been pursued with 

the rigour it demands. While many intergovernmental 

organisations and policy institutes 

do exist, which are dedicated to promoting peace, the study of 

peace-building and conflict resolution mechanisms has primarily 

been limited to the humanities community with only peripheral 

involvement from experts in data science and related areas. 

Clearly, the necessity to account for cultural nuances, human 

emotions, irrational behavior, and different notions of fairness 

in peace-building makes it a highly challenging space for a hard 

scientific approach. However, the effectiveness of data science 

approaches has been demonstrated in domains with similar 

complexity such as advertising, online political campaigning, 

personalised learning, and various forms of match-making. The 

recent Facebook-Cambridge Analytica scandal, in particular, 

highlights the potency of a data-driven approach to influence 

human behavior at a massive scale. The primary reason for 

the limited participation of the data science community 

in the pursuit of peace is that it has not has been 

sufficiently incentivised in our increasingly capitalistic 

world unlike advertising, which is now being honed to almost an 

exact science at companies such as Google and Facebook. 

This article attempts to examine certain vital elements required 

to advance peace and provides a sense of how data science can 

potentially contribute to those elements. At the same time, it lists 

caveats and external factors that need to be considered to ensure 

that the potential is realised. 

Elements of peaceful conflict resolution 

Heading home one evening in a cab, I had a close encounter 

with mob violence. A car crashed into our vehicle, and within a 

matter of minutes, a massive crowd of people assembled. Folks 

started arguing with each other, and before long, the arguments 

devolved into a violent fight resulting in multiple people getting 

injured without any resolution on handling the damages from 

the accident. This incident took me back to a more serious 

car accident that happened a few years ago in California. In 

contrast to the episode in Bangalore, the parties involved in 

the accident just pulled over, assessed the damages, exchanged 

insurance numbers, and called the authorities to lodge a FIR. We 

commiserated with each other on the unfortunate incident and 

wished each other safe journey on our way back home. Within two 

days, the insurance companies had figured out how the damages 

had to be paid for, and both parties readily accepted the decision. 

The contrast between the two incidents could not have been 

more stark. The triggers were the same, but the outcomes vastly 

different, which made me wonder why that was the case. 

There seemed to be at least four salient factors that could explain 

these differences. 

• Regulatory structures that strongly discourage violence 

• Predisposition of the population towards peaceful behaviour 

• Familiarity with the resolution process of common conflicts 

• Availability of arbiters perceived as neutral and knowledgeable 

by all the parties 

None of the above positive factors are easy to create. Neither is 

this list exhaustive as there are many other aspects (e.g., inherent 

power imbalances, malicious intent by third parties) that one needs 

to consider to address the vast variety of conflicts such as domestic 

violence, communal riots and territorial disputes. However, the 

good news is that over the last two decades, there have been 

massive advances in data management, internet technologies, 

computing technologies, and machine learning, and these 

technologies have leveraged each other to create a perfect storm, 

which makes it possible for us to attempt to solve such challenging 

problems. 

How can data science assist? 

To understand how and where data science and machine learning 

(ML) can help, let us delve deeper into the obstacles faced in 

creating some of the factors that contribute to peace. 

Regulatory structures & policy decisions. Government 

organisations and policy think tanks have enormous 

power to influence policy decisions and regulatory 

mechanisms, but often do not make good choices even with 

an abundance of good intentions. These suboptimal choices are 

primarily due to a lack of access to the relevant data as well as an 

inability to distill the available data to perform a what-if-analysis, 

e.g., is free public schooling till the age of 18 years likely to result 

ISSUE • 9 

1 9

opinion 

in fewer cases of domestic violence? 

Data science approaches can 

optimise this decision-making 

process in multiple ways by: 

• enabling semi-automated consolidation 

of relevant data from a wide range 

of heterogeneous, structured and 

unstructured multi-lingual, multi-media 

sources across the world, 

• offering insights on temporal trends, 

natural clusters, and frequent patterns to 

policymakers, 

• discovering associations between policy 

choices and outcomes for different 

contexts via analytic models learned 

from the data, 

• performing quick and accurate 

evaluation of different policy choices via 

statistical tests. 

Promoting a peaceful mindset 

among people. Today’s cultural 

forces (e.g., social media, news 

media, video games, movies) 

and education systems are much 

more geared towards promoting 

competition, aggression, and 

violence than peacefulness and 

cooperation. The current state of affairs 

most likely is due to the apathy on the 

part of the educators and media owners 

towards promoting peace. Nevertheless, 

even if a positive intent could be induced 

with changes in regulation, it is still 

non-trivial for moderators to weed out 

objectionable content and for educators 

and influencers to take appropriate 

actions to mold the behaviour of people 

under their influence since that requires 

fine-grained contextualisation. 

Data-science based systems can 

be quite helpful in solving some 

of these challenges. In particular, 

it is possible to automatically assess 

different types of content (movies, 

games, online posts) on the likely impact 

it has on a consumer, which can then 

be used to regulate the content sources. 

Twitter, Facebook, and Google already 

have such systems in place. Analysis of 

behavioural data, combined with medical, 

Today’s cultural forces 

(e.g., social media, news 

media, video games, 

movies) and education 

systems are much 

more geared towards 

promoting competition, 

aggression, and violence 

than peacefulness and 

cooperation. 

demographic, environmental factors and 

content consumption history, can also 

pinpoint the reasons for a peaceful or 

violent disposition, which can then be 

used to recommend appropriate corrective 

or learning actions to educators/leaders 

targeted towards each individual. For 

example, if a child is aggressive because 

of a lack of basic needs such as food 

and housing, it is preferable to focus on 

addressing those issues than inundating 

her/him with messages on peacefulness. 

Familiarity with common resolution 

mechanisms. A common cause for 

avoidable violence is ignorance of 

resolution mechanisms for everyday 

conflicts, e.g., domestic abuse of women 

by their spouses, communal riots, 

economic exploitation of tribal and 

rural people by others. This observation 

is particularly true for the poor and 

under-educated parts of the population. 

ML - based personalised content 

recommendation systems that can 

offer content on conflict resolution 

processes tailored to a person’s 

situation, in the language and 

media of his/her choice can go 

a long way in addressing this 

problem. One can also create engaging 

role-playing games powered by ML 

algorithms that can simulate typical 

conflict scenarios and train the players on 

navigating such situations in real world. 

Neutral, knowledgeable and 

effective arbiters. Resolution of 

conflicts is often tricky in the absence of 

a neutral, knowledgeable arbiter. In most 

conflicts, it is not only the neutrality that is 

in question; the sheer volume of disputes 

makes even the number of available 

arbiters extremely low. Automated ML 

agents by design do not have a bias 

specific to a particular situation and are 

more likely to be neutral and perceived 

as such. It is also a scalable option to 

deploy automated agents directly or as 

assistants to humans to enable thoughtful 

and timely resolution of conflicts. There 

is, in fact, ongoing work on using 

ML models to assist judges in 

coming up with objective and fair 

judgments. Specialised ML systems are 

also being used to evaluate the truth of 

objective claims from different sources 

based on the raw evidence, which can 

help in resolving differences. 

The above ideas on applications of data 

science for peaceful conflict resolution are 

just the tip of the iceberg, and there are 

possibly many more. 

Caveats on the limits of 

data science 

Intelligent ML-based systems such as 

Google search, Facebook feeds, Amazon 

product recommendations and Uber 

have become an integral part of our lives, 

but most of us do not yet have a good 

understanding of how data science and 

ML work. Misconceptions around data 

science make it seem like either like a 

“panacea” or an “out-of-control beast”, 

leading to all kinds of absurd headlines 

full of hype or fear. The reality is far 

different, and it is essential to understand 

this to ensure appropriate and effective 

use of data science. 

The core of data science is machine 

learning, which comprises techniques 

that enable machines to learn from data 

without being explicitly programmed. 

These techniques are different from the 

older generation AI methods such as rulebased 

systems that had to be explicitly 

programmed by experts. In fact, ML methods have more in 

common with the way humans learn. For example, consider how 

a child learns. To him/her, it is important to determine if a new 

person is friendly or hostile. The learning process involves making 

multiple observations over time on people’s attributes such as the 

tone of voice, facial expressions, and creating a mental model of 

hostile vs. friendly, which can be then be employed to characterize 

a new person. 

Machine learning operates similarly and involves three key 

elements. 

• Explicit representation of all the relevant information, e.g., the 

tone of voice, as variables in a mathematical framework 

• Collection of empirical observations on the variables 

• Learning algorithms to infer associations between variables using 

the gathered data. 

The benefits of data science methods relative to manual judgment 

stem from their ability to effortlessly handle trade-offs among 

multiple possibly conflicting signals, incorporate massive amounts 

of data, and contextualise predictions for fine-grained contexts. 

However, one must be extremely mindful of the fact that ML 

techniques have a heavy dependence on data and do not have 

any innate ability to generate predictions. In particular, the most 

commonly used class of ML methods, i.e., supervised learning 

methods, need adequately labelled data to first learn what they are 

expected to predict, e.g. if we want to rate content as peaceful or 

violent, we first need sufficient training examples of peaceful and 

violent content. Furthermore, the applicability of an ML model 

is often limited to data distribution it is trained on, e.g., models 

trained on the US population might or might not generalize to 

a different community. If it turns out that data used to build ML 

models is biased or noisy, then the predictions of the resulting 

models are also likely to be biased and erroneous. 

External regulatory factors needed for 

success 

Like any other powerful technology, data science is 

a sharp double-edged sword that can cut both ways. 

Consolidation of data not only unlocks valuable insights but 

also poses enormous risks if it falls into the hands of malicious 

players. It is imperative, therefore, to carefully formulate policies 

that safeguard against privacy breaches and potential misuse, and 

ensure strict compliance. The GDPR regulation is a commendable 

effort in this direction, but more work remains to be done. 

There is considerable evidence to demonstrate that 

privately-owned information ecosystems such as search 

engines, social media networks and news media, driven 

by engagement-drive business metrics, are resulting in 

increasingly polarised communities. Data science methods 

are being aggressively deployed in these ecosystems but primarily 

to optimise business metrics, which are not necessarily aligned 

towards creating a peaceful society. How much ever governmental 

and inter-governmental organisations attempt to promote peace, 

Like any other powerful technology, data 

science is a sharp double-edged sword that 

can cut both ways. 

these efforts are likely to be subverted by these highly influential 

ecosystems. It is critical to impose well-thought-out content 

standards and regulatory mechanisms that make it economically 

profitable for these privately-owned ecosystems to promote peace. 

Summing up 

To conclude, data science has the potential to make immense 

contributions to the pursuit of peace. Providing macro-level 

insights to policymakers, curating personalised content on peaceful 

conflict resolution, guiding educators in molding the behaviour of 

their wards and automated agents for arbitration are some of the 

many ways that data science can assist in promoting peace. There 

are, of course, caveats on the applicability of data science and risks 

of privacy breaches that should be taken into consideration. Most 

importantly, it is essential to harness the capitalistic 

forces of the world towards the pursuit of peace. 

OPINION 

ISSUE • 9 

2 1

opinion 

Declaration of 

Independence II 

OPINION 

DAN SHEFET 

Data Subjects of all countries unite ! 

We were promised ultimate freedom and liberty and instead we have 

become serfs of new masters. 

Your tyranny has overshadowed and transcended that of 

government- yet you are not elected. 

You have highjacked our space, our freedom and our sovereignty. 

You have no moral right to rule us or possess us. 

You are not welcome among us. 

You do not answer to democracy, but you will have to answer to 

us, the People. 

Your arrogance and indifference to human tragedy, your abuse 

and avid concentration of power and hegemony disqualify you 

from the role you have usurped. 

You have transformed our dreams and promises of decentralised 

freedom into a centralised nightmare. 

Dan Shefet: Lawyer at the Paris Court of Appeal 

(France) 

Dan Shefet is a frequent speaker at international 

conferences and academic venues on IT Law, Data 

Privacy Content Regulation and Human Rights on 

the Internet. In 2014, he founded the Association 

for Accountability and Internet Democracy (AAID), 

the main objective of which is to introduce a general 

principle of accountability on the internet. 

Under the pretext of saving the world you fill your coffers 

without contributing to society. 

You yield before no manipulation. 

You have degraded us to products in your realm. Products that 

deserve no respect. 

Governments cannot and will not stop you, but we the People will. 

We will boycott your services. We will take our clicks away from 

you and your empire will crumble. 

You have no right to make judgments affecting life and death. 

We no longer trust you. 

What you fear most is our passion. What you lack most is 

compassion. 

Privacy is not dead. Get over it. 

Now, here are our demands: 

You will give us back control over our lives. 

We will only allow you to co-exist with us if you accept that with 

power comes obligations and that your disregard of ethics and 

social responsibility will cease. 

You are not above the law and your failure to assume your 

obligations and contribute towards society are shameful. 

Your abuse of your overwhelmingly dominant position and your 

failure to take ethics into account when developing, or rather, 

acquiring new technology and services (you develop little yourself) 

are intolerable. 

You will allow Data Subjects all over the world to take 

control of their own data and immediately heed their 

demands for deletion and de-referencing. 

You will not anoint yourself to be the judge over the 

appropriateness of such demands. 

You will allow public inspection to verify that these demands 

have been met and that data has been definitely and permanently 

erased from your gigantic data troves - because we do not trust 

you. 

You will create an ethics board with independent experts who will 

monitor all of your new “technology” whether developed in-house 

or acquired and in particular those technologies that apply to 

artificial intelligence. 

You will refrain from further abuse of your dominant position 

destroying any alternative to your « services » and consolidating 

your power even more. 

This time we will create a true civilization of the Mind in 

Cyberspace. 

May it be more humane and fair than the world you have made. 

P a r i s, France 

August 3, 2018 

Dan Shefet 

ISSUE • 9 

2 3

opinion 

Dreams and Reality: 

OPINION 

How AI will Change 


Harri Ketamo, founder of Headai, senior fellow in University of Turku 

Because AI will change everything, it 

is said that AI will change the way we 

learn. This couldn’t be a more misleading 

statement. 

What is learning as a phenomenon 

Harri Ketamo, Ph.D. is an entrepreneur with 20 

years experience in learning sciences and artificial 

intelligence as science and as business. Currently 

he is founder & chairman of Headai, a jobTech 

company solving global talent mismatch and 

workforce re-skilling challenges with AI. He’s 

also actively participating in academic research 

as an adjunct professor at Tampere University of 

Technology and as a senior fellow at University of 

Turku. Previously he has been founder & CEO of 

gameMiner (game AI that learn from players), xTask 

(adaptive learning platform for mobile devices) and 

SkillPixels (elementary school as a mobile game). 

Ketamo was awarded the Eisenhower Fellowship in 

2017. 

The way people learn is an outcome of millions of years of 

evolution and it cannot be changed with technology. In short, 

we learn through interaction by observing the world around 

us i.e. we make observations, conceptualise our observations 

and connect them into our existing understanding. When we 

connect observations we have made before, we just strengthen 

our current understanding. When adding new observations, we 

learn. When we make observations that don’t fit into our existing 

understanding, we might end up learning and at the same time 

changing our understanding radically. 

In science, the learning process is studied at the for example 

chemistry level, biological level, neural cell level, social level 

and psychological level. All science supports this generalised 

definition of learning, no matter if we call it conceptual learning, 

constructivism or connectivism. Claims that AI will change the 

way we learn is just like a claim that television will change the 

way we learn. Every ten years, there have been technologies that 

have claimed to change the way learning happens. Games (2010), 

mobile devices (2000), Internet (1990), CD-rom (1980), television 

(1970), programmed instruction (1960), radio (1950) and spirit 

duplicators (1930), just to mention a few. 

We dream that technology would make learning fast and easy, but 

that’s an illusion. 

However, all technologies, including artificial intelligence, 

have changed our abilities to make observations of the world. 

Furthermore, technologies have enabled a remarkable change 

in education and teaching. Radio brought access to good quality 

educational content for millions. Internet provided access to a 

lot of information, readily available for learners. A decade later, 

there was no need to just deliver information in schools – new 

pedagogical innovations were developed. The Learning process, 

however, remained the same: we yet connected our observations 

into our mental understanding. Learning is hard work, but we 

can make it more enjoyable, accessible and effective with good 

pedagogy. Technology alone is never the answer. 

Coming back to Artificial Intelligence (AI) 

We should not spend too much time on defining what AI is. 

All machine performed activities that would require conscious 

thinking also from people, can be called Artificial Intelligence. AI 

is not dependent on which algorithms or technological frameworks 

are used. 

Furthermore, AI is old. As a science, it comes from the 50’s and it 

has been with us for more than 60 years. The first generation AIs 

were decision trees and state machines, based on programmed 

rules. In many solutions, the rules were based on scientific work 

and millions of lines of data, but the rules were programmed 

after people found the rules. In second generation AI, there are 

no pre-programmed rules and machines would learn the rules 

from the given data. In other words, the rules with which AI works 

are not programmed, they are (machine) learned. There are 

excellent use cases for both types of AI, for some cases 

programmed rules are perfect, because they are fast to 

perform, e.g. spell checking. Some other cases require 

machine learning because of the complexity of the case, 

e.g. speech recognition. The third generation is going to 

ISSUE • 9 

2 5

opinion 

be about extending the concept of machine learning. i.e. 

teaching machines case by case with natural language. 

AI can be used as an assistant for people, as a closely managed 

worker or as an autonomous system. In many cases, we use AI 

in the form of a co-worker, letting it to do easy, boring, dirty or 

dangerous tasks. Assisting AI performs numerous supporting tasks 

for professionals and so enabling them to focus on higher order 

thinking. Currently, there are very limited number of autonomous 

Alongside AI ethics, we should discuss 

data ethics: who is allowed to collect data, 

who owns the data, is the data real and 

valid, what are biases in the data, what is 

the role of manipulation, and so on. 

AIs in use, however, when we refer to AI, we tend to think 

autonomous AI. It is important to note that the ethical discussion 

around AI is very different if we talk about autonomous systems 

or systems that provide assistance by performing routine tasks. 

Alongside AI ethics, we should discuss data ethics: who 

is allowed to collect data, who owns the data, is the data 

real and valid, what are biases in the data, what is the 

role of manipulation, and so on. We have to accept we can 

not talk AI as an isolated system, it is always connected to other 

technologies and processes. 

And how AI will change education 

There is nothing new in applying AI to learning. In 1912, Edward 

Thorndike introduced his idea about The Learning Machine. The 

machine could ask questions of the learner and suggest further 

readings, if required. Because there were no modern computers 

invented at that point, The Learning Machine applied punch 

cards to run Adaptive Learning features 100 years before the 

concept of Adaptive Learning became popular in practice. 

Adaptive learning applies computational methods that fit into the 

AI-family, and so we can call it AI. There are also two generations 

of adaptive learning. Prior to the 1990s, adaptive learning was 

based on rules found in scientific research and then programmed 

into machines. Post the 90’s, real-time data was used in order to 

provide highly individual learning experiences for the learner and 

detailed learning analytics. This required other technologies 

like fast internet, big memories, own devices and 

computation power to support AI. 

There are parallel opportunities to apply AI in several educational 

tasks to either improve or speed up the process. However, the 

power of AI is not in improving old process, we have to rethink the 

processes. 

If we want to understand the major opportunities related to AI, 

we have to focus on one of the biggest global challenges we face in 

education: We don’t have enough teachers to provide education for 

everyone. This is critical in elementary education, but remarkable 

also in vocational training, higher education and lifelong learning. 

In early years of education, such as elementary education and 

primary education, the human teacher cannot be replaced. No 

way. Soft skills and transferrable skills such as critical thinking, 

problem solving, team work and communication needs interaction 

with people. However, we can provide AI-based teaching assistants 

and helpers for teachers and so enable teachers to work with 

bigger groups. This, however, requires AI-pedagogy, which we 

don’t have it yet. We need to invest a lot in pedagogical research 

in order to understand the full potential of AI to elementary 

education. 

In lifelong learning, work place training and professional 

development, the key factor is to understand what skills are needed 

in the labour markets now and in the future. Although there is 

research being undertaken, big data is not being applied and 

micro-skills are not being focused upon because there is too much 

data and complexity for people. 

For a couple of years, we have had AI assistance that reads 

through terabytes of job openings, curriculums, future forecast 

reports and labor market analysis, to build a real time research 

report on skills seeking. We could do this also with people, if we 

just could have millions of people to do this work. 

AI can build parallel models on an individual’s skills at micro 

level, it can easily show an individualised career path: direct jobs 

with person’s current skills as well as low-hanging jobs that are 

accessible after person gets couple of identified skills. Furthermore, 

when AI knows the curriculums, it can also reveal educational 

opportunities relevant for individual. 

When the skill gap is recognised, AI can construct tests and 

curate content packages for the individual. In other words, AI 

can produce online courses from identified skills and so bring 

vocational training and lifelong learning for billions of learners. 

Finally, AI won’t change the way we learn, but it can change the 

access to education, globally. 

Download #YouthWagingPeace now. 

Also available, Action Guidelines for Prevention of Violent Extremism (English and French versions) 

Details, visit : http://mgiep.unesco.org/youth-waging-peace 

ISSUE • 9

International Youth Campaign 

on Kindness for the SDGs 

1 

About 

The UNESCO MGIEP has initiated the International Youth Campaign on Kindness for the Sustainable Development Goals (SDGs). 

The purpose of this Campaign is to mobilise the world’s youth to achieve the 17 SDGs through transformative acts of kindness. The 

overall goals of the campaign are to: 

2 

• Create a positive culture of kindness where a person’s selfless act matters and to show how it can make a difference to the SDGs 

• Provide a comprehensive, independent, fun and engaging pathway for youth to demonstrate how an act of kindness is a non-trivial 

task and that being kind is “cool” 

3 

4 

5 

BE A PART OF THIS 

CAMPAIGN NOW! 

FOR DETAILS, VISIT 

http://mgiep.unesco.org/kindness 

1. Bhubaneshwar City, India: SDG 9: Innovation, Industry and Infrastructure 

2 & 3. Kolkata City, India: SDG 4: Quality Education: An educational session at two NGO 

4. Lucknow City, India: SDG 2: Zero Hunger 

5. Kolkata City, India: SDG 4: Quality Education: An educational session at two NGO’s 

6. Lucknow City, India: SDG 2: Zero Hunger: Food distribution among the underprivileged in 

Lucknow (in collaboration with the Robin Hood Army) 

6 

2 9

cover 

story 

COVER STORY 

Personalising ‘Learning’ - 

Can AI Promise 

Customised Education 

For ‘Humanity’ 

Nandini Chatterjee Singh 

Programme Specialist - Science Of Learning, Unesco Mgiep 

Raunak Jain 

Former Consultant, UNESCO MGIEP 

B 

outique or personalised 

learning is the final 

frontier of a successful 

education system. 

Recent research in neuroscience has 

demonstrated that each brain is ‘wired 

uniquely’ and thus ‘learns differently’. 

Thus far, education systems have focused 

on curricula geared towards producing a 

workforce for mass industrial deployment 

and have implemented the ‘one size fits 

all’ approach. Unfortunately, this has led 

to the following undesirable results: 

(1) Many individuals trained but unable to 

adapt to changing job requirements, 

(2) Little consideration for natural talent, 

thereby killing creativity, 

(3) Little or no love for learning, 

(4) Unhappy, frustrated youth. 

Consequently, the current education 

systems finds itself broken, insufficient 

and ineffective at meeting the demands of 

the 21 st century. 

For education to provide holistic 

development to children, it is urgent and 

necessary that learning become a fun 

The resurgence of 

Artificial Intelligence or 

AI, however, offers a ray 

of hope in achieving such 

a ‘personalised learning 

system 

and enjoyable experience. It should allow 

children to learn at their own pace, it 

should do so in ways that address their 

various strengths and weaknesses, and 

most importantly, it should prepare them 

to realise their full potential. For this 

to happen, learning needs to become 

‘personal’. While the premise is exciting 

and promising, such a personalised 

learning system has seemed utopian thus 

far. 

The resurgence of Artificial 

Intelligence or AI, however, offers 

a ray of hope in achieving such a 

‘personalised learning system’. In 

order to understand how AI can make 

this possible, we will give an overview of 

AI, how it has resurfaced, examples of its 

application, followed by how we expect AI 

to facilitate ‘personalised learning’. 

What is AI ? 

The term Artificial Intelligence or 

AI was coined by John McCarthy in 

1956, two years after the untimely 

death of Alan Turing, who came 

to be known as the father of AI. In 

1950, at a time when the first generalpurpose 

computers were being built, 

Turing was already grappling with the 

question “Can machines think?” He 

developed a hypothetical machine, 

called a Turing machine, for encrypting 

codes built to test and define Machine 

Intelligence and thus refers to computer 

programs that exhibit human-like 

ISSUE • 9 

3 1

COVER STORY 

intelligence such as logical reasoning, 

problem solving and 1 learning. 

The Turing machine was able to 

implement algorithms that could mimic 

the process of basic human decisionmaking. 

A number of algorithms were 

developed and some of them were 

implemented successfully in cracking 

secret codes during the war; however, 

many algorithms remain untested due to 

the limitations of computing capabilities. 

With recent advances in both computing 

power and algorithmic efficiency, AI 

gained traction. This trend is expected to 

accelerate as new computing technologies 

such as neural processors and quantum 

computing achieve stability and maturity. 

AI technology is projected to dramatically 

change the types of tasks performed 

by humans and those performed by 

machines. As a consequence, AI has 

caught the imagination of not only 

technologists and entrepreneurs, but 

also educationists and policy makers. AI 

is unique in being a generic umbrella 

technology that can enable many different 

disruptive technologies and solutions. 

Creating a Framework for 

AI 

The conception of an AI system indicates 

that it should possess the ability to interact 

with the real world. To understand AI’s 

progress thus far, we will elaborate on the 

four dimensions identified by DARPA 

( 2 Defence Advanced Research Projects 

Agency) to understand the evolution of 

AI. Based on this, intelligence may be 

broadly be 3 described. 

• Perception – The ability to use 

sensory systems and language to obtain 

information about the natural world 

• Problem analysis – The ability to process 

the information obtained above in order 

to identify a problem, analyse it and 

provide a solution 

• Abstract knowledge – The ability to 

abstract knowledge from perception in 

one domain and apply it to another 

• Experiential learning – The ability 

to continually learn from real-world 

data patterns and refine perception and 

knowledge 

Next, we will review the evolution of AI 

systems within the framework described 

above. A historical review of the evolution 

of AI systems reveals that they appear to 

have followed a different process – one 

that has been driven by algorithm and 

technology development. Based on their 

capabilities, developed AI systems have 

been categorised in distinct generations. 

Gen 1 AI – Rule-driven 

Learning 

The first generation of AI programmes 

described knowledge through the 

definition of static rules. One of the 

earliest programmes in this series was 

designed to played chess. Computer 

systems used a set of defined rules to play 

chess with humans and their algorithms 

became so good that the systems even 

beat humans. Such systems offered limited 

perception of the outside world, namely 

that of the chessboard. 

Global Positioning Systems (GPS) 

are a second example of firstgeneration 

AI. They demonstrate 

the ability to identify location, 

analyse and contextualise it on a 

map, and provide solutions in the 

form of directions. GPS systems 

possess very limited abilities to cope 

with deviations from the routine, yet 

their presence and utility in our daily 

lives cannot be denied or ignored. 

When viewed through the dimensions 

of intelligence described above, firstgeneration 

AI systems are only capable 

of using ‘problem analysis’ for limited 

applications of the outside world. For 

instance, users of Google Maps will know 

that based on a set of pre-defined rules, a 

map prescribes a specific route. However, 

based on their experience, a driver may 

take a different route each day, while the 

application is unable to learn the route 

taken by the driver. Thus, while firstgeneration 

AI computer programs possess 

great capabilities and are constantly being 

deployed in many industries for the value 

that they provide, they are capable of only 

implementing rules and are unable to 

learn from patterns and implement new 

trajectories or outputs. 

Gen 2 AI – Data-driven 

Learning 

The next big leap in AI systems happened 

with the advent of Big Data. For a 

long time, access to limited data, which 

was isolated and held in data cocoons, 

prevented AI from being tested or 

implemented in demographic settings. 

Recent advances in data collection 

and assimilation techniques have made 

available to us enormous amounts of 

data, which is analysed using advanced 

data analytics methods – this entire 

process is referred to as Big Data. Further, 

the migration from regular CPUs to 

extremely powerful computing devices 

has enabled the development of parallel 

intelligence engines. AI programmes have 

leveraged this to great advantage; for 

instance, novel transformative algorithms 

are now available to identify clusters 

and patterns in datasets. Often, patterns 

not visible to humans can be located 

using such algorithms, sometimes even 

through trial and error. Thus, training 

a system using Big Data has made a 

huge impact on utilising the power of 

‘pattern recognition’. However, it is brittle 

– it breaks as soon as soon the data set 

changes. 

One example of second-generation AI’s 

implementation is Microsoft’s Twitter 

Bot 9 and the controversy surrounding it. 

Popularly named Tay, this teen-talking AI 

chatbot was built to mimic and converse 

with users in real time; it was released via 

Twitter in 2016. As it began to converse 

with users, some people took advantage 

of Tay’s machine learning capabilities 

and began to teach it racist and offensive 

language. As a consequence, the bot 

began to post offensive tweets through its 

Twitter account, forcing Microsoft to shut 

down the service only 16 hours after its 

launch. 

Categories of 


Gen 1 AI – 

Rule Driven Learning 

The first generation of AI programs that were 

developed incorporated computer programs that 

described knowledge through the definition of 

static rules. One of the earliest in this series, was 

the development of computer programs that 

played chess. Such systems offered limited 

perception of the outside world, namely that of 

the chessboard. 

Gen 2 AI – 

Data Driven Learning 

The next big leap in AI systems happened with 

the collection and availability of BigData. For a 

long time, access to small limited data, which 

was isolated and held in data cocoons, prevented 

AI from being tested or implemented in 

population settings 

Gen 3 AI – 

Context Driven Learning 

Building from the earlier paragraph, it this 

therefore quite clear that the primary focus of 

3rd Gen AI is to build and incorporate algorithms 

that understand context and implement an 

analytical path to decide response. 

COVER STORY 

ISSUE • 9 

1 http://cpsc.yale.edu/research/artificial-intelligence 

2 

https://www.darpa.mil/about-us/darpa-perspective-on-ai 

3 

https://www.iata.org/publications/Documents/AI-White-Paper.pdf 

3 3

COVER STORY 

While the incorporation of Big Data 

offers tremendous promise and is exciting, 

second-generation AI lacks the power of 

logical reasoning, understanding context 

and abstracting knowledge to different 

domains. Human beings learn in multiple 

ways – through experience, patterns, 

and imitation – to list a few. Clearly, 

such ideas needed to be translated to 

algorithms and incorporated in thirdgeneration 

AI. 

it would need to understand context 

and have the capability to go from one 

domain to another; abstracting knowledge 

and learning. Thus, the possibility of a 

system or machine understanding the 

context of its environment, and learning 

and adapting based on changes in that 

environment, is possibly the next frontier 

of AI. 

The Role of AI in Education 

To understand the role of AI 

in education, we first have 

to understand the concepts 

which make education 

systems successful, which 

will then help us understand 

how AI can add value to 

each component in the 

system. 

COVER STORY 

Gen 3 AI – Context-driven 

Analysis 

Building upon our argument, it is quite 

clear that the primary focus of Gen 3 AI 

is to build and incorporate algorithms 

that understand context and implement 

an analytical path to decide response. As 

discussed earlier, the data sets on which AI 

systems are trained are brittle. Thus, the 

Gen 2 AI is only as good as its data set it is 

trained on and is incapable of recognising 

right from wrong. For instance, as seen 

in the case of Tay, when the bot was 

provided new patterns by Twitter users, 

it learnt new patterns, not realising that 

these were inappropriate patterns or even 

different from what it was taught earlier. 

This brings to the fore the value of logical 

reasoning and analysis in AI systems. 

The second aspect that Gen 3 AI needs 

to incorporate is recognising patterns 

across domains and abstractions. Humans 

beings’ ability to not only recognise 

patterns but abstract knowledge 

such that it can be used in different 

settings is incredibly powerful and 

is learnt and accomplished without too 

much data and quite early in life. One 

simple example is our ability to recognise 

handwriting, which varies hugely among 

people. Yet, the human brain is capable 

of recognising a word written in various 

ways. One faculty that the human brain 

uses here is the understanding of context, 

which is a vital aspect the way we make 

decisions. For AI systems to be more 

effective, it is important that they be 

able to take decisions based on context. 

If third-generation AI is to be effective, 

To understand the role of AI 

in education, we must first 

understand the concepts that make 

education systems successful, 

which will then help us understand 

how AI can add value to each 

component in the system. The two 

primary components of any education 

system are the student and the teacher. 

An ideal education system should provide 

content that is contextually relevant, 

technically sound, and emotionally 

rewarding to ensure that students acquires 

skills and continue to remain motivated 

to 4 learn. This can be achieved if they 

are provided an educational experience 

that is immersive, multi-sensory, paced 

appropriately, and rewarding. 

For teachers to effectively 

implement good teaching practices, 

education also needs to be engaging 

and fun, and most importantly, 

allow the teacher to continuously 

measure learning and receive 

transparent feedback. 

Artificial Intelligence embedded in 

digital learning ecosystems presents 

the possibility of achieving both of the 

above in a measurable and scientific 

manner. It is the theoretical basis on 

which personalised learning and analytics 

work to enable richer understanding 

of learning data and improve learning 

effectiveness at scale. 

The integration of AI into learning 

systems is possible because of digitisation. 

Here, digitisation refers to the constant 

interaction of the individual with the 

machine during learning, which makes it 

possible for us to collect data on decision 

making. This data would be supremely 

valuable to understand students’ learning 

patterns, types and pace, which in turn 

could be used to build and train AI 

systems. 

Imagine a digital platform presented to 

children as an interface where they can 

speak, touch and visualise information, 

and engage with it. This information 

can be provided in visual or auditory 

modes, or even as a short movie. Using 

the powerful modality of a story, children 

can be engaged in a learning experience 

that requires interaction with a digital 

platform. The story itself is delivered in 

a format that immerses children in an 

experience where they respond, interact, 

think critically and take decisions towards 

specific outcomes, while their interactions 

are recorded as a digital transaction. 

Navigating through this story would 

require children to assume a different 

identity – a common aspect of gaming 

– and then make a choice. Throughout 

this process, children are provided 

information and learn, making both 

intellectual and emotional decisions, and 

constantly questioning and reflecting on 

the situations presented in the story. Such 

a process would not only allow them to 

learn in a fun and engaging manner but 

also permit continuous assessment of the 

different skills that need to be acquired. 

The digital transactions that learners 

would undertake on such a platform 

–from conversations to assessment 

exercises to games – would have a digital 

footprint that can be used to profile every user. Central to this 

digital curriculum would be the inclusion of skills that emphasise 

compassion, collaboration and co-operation among fellow 

participants, so that children build emotional intelligence along 

with intellectual intelligence. 

For the teacher, this technology would facilitate intelligent tutoring. 

The teacher would be able to analyse each child’s learning path 

and actively generate/select content that intervenes to improve 

weaknesses, and design well-informed strategies that can be tested 

and improved in a self-learning environment. 

Over a period of time, the learning patterns and teacher 

interventions would begin to contribute to a collective intelligence, 

both from student and teacher, and then be automated by AI to 

multiple languages that would be available across classrooms. 

Here, children would also benefit from the best pedagogical 

practices that would have been gleaned from multiple teachers. AI 

would also have learnt patterns from the collective intelligence and 

thus the data set being brittle would cease to be a concern. 

Such a platform has been built by the UNESCO MGIEP, titled 

CHI (Collective Human Intelligence - chi.buzz). This learning 

platform will provide a safe learning environment for learners and 

its first version was launched at TECH 2018. 

We have not yet determined all the skills that today’s pre-schoolers 

will need in the workplace of 2030. We are hopeful, however, 

that such interactive platforms will cater to individual needs and 

provide ‘personalised learning’ to develop skill sets that will allow 

every child an equal opportunity to realise their full potential. 

ISSUE • 9 

4 

https://www.cisco.com/c/dam/en_us/solutions/industries/docs/education/Multimodal-Learning-Through-Media.pdf 

3 5

FEATURE 

ARTICLE 


CHI – A Knowledge 


Sharing Digital Platform 

Powered by Machine 

Learning 

Roy Saurabh, Chief Technology Officer, UNESCO MGIEP 

A 

rtificial Intelligence (AI) and the 

Neurosciences have been interlinked since the 

time the field of AI was at a nascent stage. 

Parallels were drawn between artificial 

intelligence and human intelligence, as early as 1956, in 

a workshop held on the campus of Dartmouth College [1]. Many 

of the early pioneers straddled both fields, with collaborations 

between the disciplines of neuroscience and artificial intelligence 

proving to be highly productive (Churchland and Sejnowski, 1988, 

Hebb, 1949, Hinton et al., 1986, Hopfield, 1982, McCulloch 

and Pitts, 1943, Turing, 1950) [2]. Whether it is Neural Network, 

Reinforcement Learning or the Machine Learning training 

Roy Saurabh has over 13 years global experience in Technology & Strategy 

Consulting. Roy has a proven track record in large-scale project management, 

Entrepreneurship, and Technology Consulting. Previously, he was the Head of 

Technology for the National Skill Development Mission, impacting more than 3 

million lives, Director for Oracle’s leading Europe Partner, Cedar Consulting UK, 

and Founder/CEO of Zosher. Roy has in-depth knowledge of different client/server 

architectures based on various frameworks, data visualisation tools, classification 

& regression models and other popular as well as advanced Artificial Intelligence 

/ NLP and Machine Learning algorithms. 

models, data scientists always have had a superior human 

intelligence counterpart to refer to while envisioning or drawing 

inspiration for AI models. 

What makes this co-existence exciting is that both fields 

have some highly specialised ‘knowns’ and ‘unknowns’. 

When we talk about blind spots in the neurosciences, a basic 

in-depth understanding of creativity, imagination, dreams 

and consciousness is still lacking. As a recent DARPA project 

study[3] indicates, while AI has taken strides in some super 

specialised pockets, basic human common sense is absent in 

at least the current applications of AI across various fields. 

With this acknowledgment, we tried to leverage the strengths 

of both the research fields of AI and the Neurosciences in our 

conceptualisation process to work towards helping leaners Learn. 

UNESCO MGIEP’s work in the field of Differential Learning 

and rich data sets that have been curated for advanced model 

training with years of education related legacy data, has helped 

us organically draw learnings from one field to the other. One 

of the major learnings of such a brainstorming session led to 

the inception of a unique knowledge sharing platform, titled 

Collective Human Intelligence (We prefer the 

Greek pronunciation /kaɪ to the Kung Fu Panda3 version). 

CHI, quite simply, is an AI-powered digital platform 

that helps extract learnings from tasks that AI does well, 

to help teach and consequently learn. CHI helps curriculum 

designers, policy makers, content developers, teachers and students 

rethink how knowledge and intelligence can be imparted in 

the 21 st century. Central to the design and development of the 

platform was the question ‘how learners learn’. 

Before delving into the technical vision and features of CHI, 

understanding the ‘WHY’ behind the development of CHI is 

critical. The two factors that fueled CHI’s conceptualisation were: 

• Limitations that the current content creation mainstream 

platforms present, both in the form of content design tools/ 

content delivery modes and the limited data insights that existing 

platforms provide for the creator community. 

• Ownership of the data generated by existing platforms solely lies 

in the hands of for-profit organisations, depriving the world of the 

invaluable data insights that could help take corrective measures 

for the betterment of the world. 

These limitations, when thought of in the context of education 

and learning outcomes, become even more critical because of the 

dynamic nature of today’s education ecosystem. Since, we are 

trying to prepare learners for such dynamic environments; creating 

engaging learner tools. In many ways, this puts into practice 

Howard Gardner’s multiple intelligences pedagogical framework 

[4]. 

When the reasons became compelling enough, the technology 

requirements and long-term vision began taking shape. CHI uses 

the latest innovative technology to ensure maximum 

engagement of creators and learners as well as provide 

deep insights from the crowdsourced learner & creator 

datasets. The scale of the impact and the number of data points 

per platform-registered user are a dream for any Technology team 

ISSUE • 9 

3 7


trying to do something meaningful with Artificial Intelligence 

(Deep Learning in particular). With rich Datasets to train on, 

the key features that we are setting out within CHI are Machine 

Translation [5], Sequence Tagging, Sentiment & Emotion 

Analysis, Recommendation Engine and Handwriting Recognition. 

These features, when implemented in conjunction with learnercentric 

platform design, would slowly start to unleash the power 

of technology in the educational context. The platform, in the 

first phase, tries to solve real issues that are left unaddressed in the 

legacy platforms from both learners’ & creators’ perspectives. 

From the content creator’s point of view, CHI offers rich 

content creation abilities, including embedding audio, 

rich text, video, journaling, games (developed using 

Unity/Phaser) to name a few. The platform framework 

design has been made extremely intuitive and powerful to help 

significantly reduce creators’ learning curve. Creators are further 

armed with real-time emotional analysis insights related to course 

reviews, discussions and facilitated dialogues. These insights help 

creators identify the temotion shift on pertinent topics and identify 

influencers by studying context-specific trends. Further, with a 

powerful character recognition feature, CHI helps content creators 

in uploading and transcribing offline legacy content, thus reducing 

the data entry effort drastically. Finally, advanced analytics helps 

creators be mindful of what is working and what is not in terms 

of the learning outcomes of the respective courses. Creators 

have the ability to ‘create’ customised learner-centric courses by 

personalising features such as the layout of the learning blocks 

or button design. Simultaneously, they have access to advanced 

analytics such as student dropout rate and time spent on lessons, 

modules or the courses. Predictive analytics within CHI can 

potentially help creators take pro-active actions, especially in the 

case of at-risk learners who may be on the verge of dropping out. 

We hope that these power-packed features of CHI are leveraged 

at scale, helping embed global best practices for future content 

creators. 

The real strength of CHI, however, lies with the various features 

available for learners; the development of CHI was conceptualised 

with learners at the center. Learners experience a guided yet 

flexible and a highly personalised learning environment. Every 

interaction on the platform is utilised as a way to learn more about 

the learning habits of the learner to help achieve the learning 

outcomes using the best available digital pedagogical learning 

mode within CHI. 

Further, deeply embedded analytics and user behaviour indicators 

of learners as well as collaborative-based recommendation 

engines help in providing recommendations of the most suitable 

learning resources available from the CHI repository for learners, 

in areas where they need maximum support. Learners have an 

opportunity to experience the immersive courses in the form 

that they are most accustomed to or adept at, without diluting 

the course’s learning outcomes. With contextual metadata, CHI 

facilitates instant feedback, review as well as AI powered help 

facilitating trial & error learning in a judgement-free environment. 

CHI leverages AI for adapting to the pace of learning and can 

consistently offer more complex tasks suited to the pace of the 

learners, making learning completely non-linear. Finally, learners 

are provided an opportunity to experience unique content (such 

as course material on SEL developed by UNESCO MGIEP) that 

would be exclusive to CHI. 

We acknowledge that the sheer mention of Artificial Intelligence 

or AI in various fields invokes mixed reactions and feelings. Some 

see automation in general and AI in particular as a threat to 

current and future jobs. Few subscribe to even more extreme views 

quoting Elon Musk’s dire warning about AI[6] in every context 

that even vaguely mentions automation. Then there are some, who 

are downright excited about AI as the long-awaited game changer 

that is bound to make us more human by automating everything 

else that doesn’t fit the definition of our innate ‘humanness’. At the 

end of the day, it does not matter which end of the spectrum you 

fall under on the topic of AI, one thing that no one cannot deny 

is that Artificial Intelligence has already made its way in almost all 

sectors. Therefore, in our view, one of the first things that requires 

our immediate attention is the role of digital technology in field 

of education. While for most of us, our education systems did not 

or could not prepare us for the Fourth Industrial Revolution or 

what we would prefer to call as the First Intelligence Revolution, it 

would be remiss for us to make the same mistake by not preparing 

our future generations for a change that is imminent. 

With so much to offer to all stakeholders, the long-term vision 

requires patronage from the Application Development community. 

CHI reach its true potential only when supported by a global 

community of learners, content developers and application 

developers. We believe that a community committed to learning 

will ensure CHI stays true to that vision; embedded in that vision 

is a truly open-source, community-supported platform that gains 

from the Collective Human Intelligence. It is no doubt a grand 

experiment but one that needs to be explored and implemented if 

we truly believe in transforming education for humanity! 

References 

[1] https://pdfs.semanticscholar.org/d486/9863b5da0fa4ff- 

5707fa972c6e1dc92474f6.pdf 

[2] https://www.cell.com/neuron/fulltext/S0896-6273(17)30509- 

3 

[3] https://techcrunch.com/2018/10/11/darpa-wants-to-teachand-test-common-sense-for-ai/ 

[4] https://www.niu.edu/facdev/_pdf/guide/learning/howard_ 

gardner_theory_multiple_intelligences.pdf 

[5] https://arxiv.org/pdf/1808.03867.pdf 

[6] https://www.cnbc.com/2018/03/13/elon-musk-at-sxsw-a-i-ismore-dangerous-than-nuclear-weapons.html 

ISSUE • 9 

3 9

interview 

Interview with 

can also be of immense help in areas such as teacher performance 

analysis and teacher placements. In higher education, data science 

can help in student on-boarding, curriculum design and student 

lifecycle management. 

AI can also facilitate experiential learning by simulating study tours, 

which are typically only undertaken by students living in urban 

areas. Videos, augmented reality and virtual reality can make this 

happen, for instance, by simulating historic monuments, landforms, 

motor mechanics, and more. 

INTERVIEW 

Dr. Avik Sarkar, 

Head – Data Analytics Cell 

at NITI Aayog, Govt. of India 

An Interview with Akriti Mehra 

3. What do you perceive as threats of AI in 

education / how can these be addressed? 

Teachers are the most important factor in ensuring the success of 

education. The most successful universities are the ones with the 

best teachers. In the context of schools, the countries with the best 

educational outcomes have a rigorous process to select teachers and 

reward them for imparting quality education. Replacing teachers 

with AI-based adaptive learning robots could lead to low learning 

outcomes in students. The education sector should be wary of 

replacing teachers; instead, it should use AI to complement 

teachers’ efforts in imparting quality education. 

6. What is the current landscape of AI and 

education in India in terms of the no. of companies 

/ institutions working on this? 

Several e-learning platforms are gaining popularity in India. 

There is a huge opportunity for these e-learning platforms to 

help fill students’ learning gaps by explaining concepts in a 

simplified manner through the use of digital technologies. The 

Government of India has also launched its e-learning platform to 

enhance the learning process for students. 

1. How will AI impact the future of education? 

Artificial Intelligence or AI provides predictions on new scenarios 

based on learnings from large volumes of historical data. We have 

to take a holistic look at a range of digital transformation in the 

education sector, along with aspects such as early prediction of 

student marks. Augmented reality, virtual reality and videos can 

provide great help to students in visualising various letarning items. 

Blockchain can be used to validate degree certificates assigned to 

students. Hence, we should look at range of digital technologies 

rather than focusing solely on AI in the education sector. 

Adaptive self-learning is the most popular AI application for the 

use of data, analytics, and AI for governance, policy-making and 

inclusive growth in India. Again, this would require personalised 

PCs/laptops – how many students in India can afford this? Hence, 

adaptive self-learning can be possibly suggested for higher studies 

or while preparing for entrance exams. 

Dr Avik Sarkar is an Officer on Special Duty at NITI Aayog, a premier policy thinktank 

of the Government of India. Dr Sarkar is heading the Data Analytics Cell at NITI 

Aayog, developing a roadmap for the use of data, analytics, and AI for governance, 

policy-making and inclusive growth in India. Dr Sarkar has over 17 years of experience 

across different aspects of data analytics, statistical modeling, data/text mining across 

companies such as IBM, Accenture, Nokia, NASA, Persistent Systems, Zycus, etc. 

He has been nominated among the “Top 10 Data Scientists in India” in 2017 by the 

Analytics India Magazine. 

2. How can AI, big data and data science be used 

effectively in education? 

We need to take a step back and examine what data is 

available from India’s education sector to enable data science 

and related activities. Once data is available, it can be used for 

discovering patterns or predicting insights from data. For example, 

in India, while information on schools is collected, data on student 

performance is hardly available in the digital format. This results 

in limited potential for analysis. Access to student performance 

and other student-related data in the digital format can help 

predict future scores and potential drop-outs. 

Nevertheless, the DISE (District Information System for Education) 

has excellent data on school infrastructure, facilities, etc. Data 

science can help understand patterns that are most relevant to 

predict student drop-outs and low learning outcomes in schools. It 

4. How will the role of teachers / educators 

be affected with the permeation of digital 

technologies and AI? 

AI and other digital technologies should be used to help 

teachers in imparting education. They can be used for predicting 

student potential and marks as well as chances of at-risk students 

dropping out. This information would be valuable for teachers in 

targeting specific interventions for at-risk students. The state of 

Andhra Pradesh had conducted a pilot to predict drop-outs based 

on past student scores and student backgrounds. In one of my 

previous roles, I had worked towards predicting students’ scores at 

Singapore schools, based on historic performance, extra-curricular 

activities and socio-economic data. 

AI and allied digital technologies can also be used to mark 

students’ answer sheets (including hand-written ones), thereby 

facilitating noting each student’s strengths and area of 

improvement. This will save enormous time for teachers by 

allowing them to document detailed personalised feedback for 

each student. 

5. What is the role of data and artificial intelligence 

in India’s inclusive growth? 

In cases where teachers are not adequately skilled, AI-based 

technologies can help them identify areas of improvement and gain 

new skills. AI can be a major tool in ensuring inclusive education 

if it is used to identify students who need special teacher 

intervention. 

7. What are some of the limitations / challenges that 

you foresee with regard to the implementation? 

These digital technologies would require personalised digital tools 

such as computers/laptops/tablets, along with stable internet 

connectivity. As mentioned previously, digital access would remain 

a major challenge as many rural areas do not yet have access to 

quality internet connection or personalised digital devices. Great 

progress is being made by the government to bridge the learning 

gaps and reap the benefits of digital technologies. 

ISSUE • 9 

4 1

11 catalytic sessions with over 50 speakers 

THEME 01 

THEME 02 

THEME 03 

THEME 04 

THEME 05 

Key Updates 

at TECH 2018 

Transformative 

Gaming 

and Digital 

Pedagogies for 

Beyond Four Walls 

of The Classroom 

Transformative 

Gaming 

and Digital 

Pedagogies for 

Artificial 

Intelligence and 

the Future of 


The Institutional 

Framework for 

Application 

of Digital 

SEL 

STEM+ 

Technologies 

THEME 

in Education: 

Towards 

TECH 2018, UNESCO MGIEP’s international conference, organised with the State Government of Andhra Pradesh, showcased the role 

of games and digital learning in enabling a shift from “transmissive pedagogies” to “transformative pedagogies” to create peaceful 

and sustainable societies. A key output of the conference was the adoption of the Vizag Declaration on Guidelines for Digital Learning 

as well as the signing of several Memorandums of Understanding (MOUs) and launch announcements. 

Surveillance or 

Collaborative 

Intelligence? 

mgiep.unesco.org/tech2018 

DATES | November 15 – 17, 2018 

VENUE | Novotel, Vizag, Andhra Pradesh, India 

KEYNOTE SPEAKERS 

Jessica Lindl – Global 

Head of Education at Unity 

Technologies. 

Dan Shefet – Lawyer at 

the Paris Court of Appeal 

(France). 

Gregoire Borst – 

Professor of Developmental 

Psychology and Cognitive 

Neuroscience of Education, 

LaPsyDE. 

Peter Vesterbacka – 

Co-founder at Lightneer, 

Fun Academy. Formerly 

Mighty Eagle at Angry 

Birds. 

Harri Ketamo – Founder 

and Chairman, Headai 

50+ BREAKOUT 

SESSIONS 

TECH TALK – TECH Talk 

by Anantha Duraiappah, 

Director, UNESCO MGIEP 

DISRUPTOR’S 

PANEL - Artificial 

Intelligence in 

Education: Who 

Owns and Who 

Manages 

4 3

YOUTH VOICES 

Can Artificial Intelligence 

Help Us Achieve Universal 

Quality Education? 

JAMES HODSON 

James Hodson is a researcher, entrepreneur, and social activist. His work spans Machine Learning, 

Economics, Sustainable Global Development, Sociology, and Philosophy. James currently serves 

as CEO of the AI for Good Foundation; and as Senior Researcher at the Artificial Intelligence 

Department of the Jozef Stefan Institute, Slovenia. Through these roles, James seeks to effect 

lasting change in our economies, societies, and scientific understanding. 

Improving Social 

Accountability of 

Government Purchases 

Through Financial 

Statements Data Mining 

In recent work from Brazil’s Office of 

the Comptroller General, researcher 

Rommel Carvalho found that even a 

simple implementation of an analytical 

platform for government transactionlevel 

spending could engender more 

responsible fiscal behaviour from 

officials. The project automatically 

identified instances of spending 

anomalies in procurement systems 

(e.g., hundreds to thousands of dollars 

for a can of Coca Cola), and allowed 

overseers to identify and incentivise good 

behaviour. 

OECD data shows quite clearly a relationship between increased 

per capita expenditure on education, and increased per capita 

GDP. Intuitively, more education is linked to stronger economies, 

stronger institutions, and stronger societies. However, spending 

more doesn’t necessarily translate directly into better education 

outcomes, and countries with higher perceptions of corruption 

tend to fare worse, since the money that is available is not used 

efficiently. Corruption, transparency, and accountability are areas 

where Artificial Intelligence can have an important role, ensuring 

capital flows to the right places and is able to impact the lives of 

the end users--the students. 

Being inclusive doesn’t just mean solving corrupt 

practices in terms of monetary flows or systemic 

biases. It also means tailoring our education systems to 

individual, emotional, cultural, and societal needs. These 

types of adaptation require data and knowledge about individuals- 

-data that must be protected and used only with the understanding 

and consent of the humans involved. Privacy, a key factor here, is 

yet another emerging area of research in Artificial Intelligence--to 

ensure we bring the benefits of the data revolution to everyone, 

without jeopardising trust. 

YOUTH VOICES 

T 

here is hope and expectation that the 

maturing of Artificial Intelligence (AI) 

techniques over the past years could lead to 

significant improvements across the social 

spectrum. From improving global health and 

well-being to alleviating food security concerns, preventing crime, 

and addressing corruption, AI has been heralded as a universal 

solution. At the AI for Good Foundation, we measure the potential 

for Artificial Intelligence to have an impact to achieve the Sustainable 

Development Goals. One of these goals is to ensure quality 

and inclusive education for all, and promote lifelong learning. 

However, learning has been one of the areas least impacted by 

the onslaught of data science and machine learning--with little to 

speak of beyond light personalisation of Massive Open Online 

Courses (MOOCs). 

It would be unfair to say that no effort is being dedicated to this 

task. Certainly, there are researchers working on virtual 

tutors 1 , automated text-book creation 2 , personalised 

learning paths 3 and more 4 . However little as yet has made it 

into mainstream adoption, and even less has been shown to have 

consistent positive impact on learning outcomes. Classrooms, 

training and continuous education in developed countries (those 

with ample resources to experiment) look more or less as they did 

50 years ago, except with access to the internet, and perhaps some 

“educational activities” on computers that do little more than 

distract classrooms and disrupt social learning. 

What can and should we do to bring the benefits of big 

data and Artificial Intelligence to improve quality of 

learning? 

More Open Data, More Open Education 

Resources 

In online settings, massive amounts of data are being collected 

regarding learner choices, behaviours, and outcomes. Some 

of these data are being used to make recommendations, or 

to improve courses in future. These data need to be properly 

harnessed to drive our understanding of how individual students 

learn best, and to give them the guidance and resources they need 

on a case by case, self-serve basis. Learning settings need to escape 

the classroom into the real world--allowing people to continue 

learning along their personalised paths to their learning goals at 

any time. 

Intelligent, Supportive Learning 

Environments 

In classroom settings, teachers often do not have the 

time for highly personalised 1-1 tutoring and feedback. 

Instead, work is set to be reviewed at a later time, introducing a lag 

between effort and feedback. This separation dilutes the learning 

experience, and makes it harder for teachers to tailor content to 

their students. Classroom support “AI Tutor” systems have been 

available since the 1970s’, and yet today’s most sophisticated 

learning technologies (e.g., Rosetta Stone’s language learning 

software) use only very basic machine learning (for example, to 

match speech patterns) to help students succeed. Fifty years on, 

learning environments should be immersive, interactive, and 

virtual tutors should be there for students whenever a teacher 

cannot be, promoting social, emotional and conceptual learning. 

Promote Economic Opportunity 

Understanding the relationship between learning, skills, and 

employment is critical for the future strength of our economy. 

Artificial Intelligence can help us to identify trends from 

local to global scales, and prepare the workforce of 

tomorrow today. Not only can a detailed view of the current 

mosaic of supply and demand of abilities help us to ensure 

consistent high quality employment, but also to appropriately 

incentivise the learning of skills that will create economic 

opportunity in future. Labour economists and AI researchers can 

work hand in hand to ensure appropriate models are built and 

mechanisms derived to extract maximum social surplus. 

Even the best systems are useless if they don’t reach the 

people who need them most. Can AI help us to improve 

global access to learning? 

Eradicate Corruption 

Education is the cornerstone of human prosperity and 

flourishing. Education brings understanding, compassion, and 

the means to make a positive difference in the world. Now is 

the time to bring the benefits of technology, data and 

Artificial Intelligence to learning at all levels, in all 

environments and for all people. With focused efforts today, 

we can break down the barriers, inequities and misunderstandings 

that prevent us from achieving our full potential. Artificial 

Intelligence is not a one-stop solution, but its presence can 

allow us to place renewed emphasis on the identification and 

implementation of new paradigms for learning and development. 

Let’s use this as an opportunity to take big steps forward. 

About the AI for Good Foundation 

The AI for Good Foundation is a US 501(c)3 Public Charity and 

global non-profit that fosters activities to maximise the benefit 

of AI technologies for social good through the lens of global 

sustainable development. Through a blend of foundational 

research, outreach, education and policy engagement our current 

projects are helping to advance the global sustainable development 

agenda — a set of goals adopted by 150 countries to end poverty, 

protect the planet, and ensure prosperity for all. Find out more 

about the AI for Good Foundation and how you can help by 

visiting https://www.ai4good.org/. 

ISSUE • 9 

1. 

Johnson, W.L. & Lester, J.C. “Face-to-Face Interaction with Pedagogical Agents, Twenty Years Later”, 

Int J Artif Intell Educ (2016) 26: 25. https://doi.org/10.1007/s40593-015-0065-9 

2. 

Singh, Rohit, Sumit Gulwani, and Sriram K. Rajamani. “Automatically Generating Algebra Problems.” AAAI. 2012. 

3. 

Chen, Chih-Ming. “Intelligent web-based learning system with personalised learning path guidance.” Computers & Education 51.2 (2008): 787-814. 

4. 

See, for example, “Artificial Intelligence in Education”, C.-K. Looi et. al. (Eds.), IOS Press (2005). 4 5

YOUTH VOICES 

YOUTH VOICES 

REFERENCES 

“AI for Good Foundation.” AI for Good Foundation, ai4good.org/. 

Carvalho, Rommel N., et al. “Probabilistic Ontology and Knowledge Fusion for Procurement 

Fraud Detection in Brazil.” Uncertainty Reasoning for the Semantic Web II Lecture Notes in 

Computer Science, 2013, pp. 19–40., doi:10.1007/978-3-642-35975-0_2. 

Carvalho, Rommel, et al. “Using Clustering and Text Mining to Create a Reference Price 

Database.” Learning and Nonlinear Models, vol. 12, no. 1, 2014, pp. 38–52., doi:10.21528/ 

lnlm-vol12-no1-art3. 

Chen, Chih-Ming. “Intelligent Web-Based Learning System with Personalised Learning Path 

Guidance.” Computers & Education, vol. 51, no. 2, 2008, pp. 787–814., doi:10.1016/j. 

compedu.2007.08.004. 

Fairfield, John. “Speech Comparison in the Rosetta Stone” Proceedings of a Symposium on 

Computer Mediated Language Assessment and Evaluation in Natural Language Processing - 

ASSESSEVALNLP ‘99, 1999, doi:10.3115/1598834.1598837. 

Farzindar, Atefeh, and Kešelj Vlado. Advances in Artificial Intelligence: 23rd Canadian 

Conference on Artificial Intelligence, Canadian AI 2010, Ottawa, Canada, May 31 - June 2, 

2010: Proceedings. Springer, 2010. 

“Home .:. Sustainable Development Knowledge Platform.” United Nations, United Nations, 

sustainabledevelopment.un.org/. 

Hoppe, Ulrich. Artificial Intelligence in Education: Shaping the Future of Learning through 

Intelligent Technologies. IOS Press, 2003. 

Johnson, W. Lewis, and James C. Lester. “Face-to-Face Interaction with Pedagogical Agents, 

Twenty Years Later.” International Journal of Artificial Intelligence in Education, vol. 26, no. 

1, 2015, pp. 25–36., doi:10.1007/s40593-015-0065-9. 

Lee, Fong-Lok. “Web Algebra Tutor: A Web-Based AI-Enabled Interactive Learning 

Environment in Algebra.” The International Journal of Learning: Annual Review, vol. 8, no. 

1, 2004, doi:10.18848/1447-9494/cgp/v08/47970. 

Li, Tao, and Samuel Sambasivam. “Automatically Generating Questions in Multiple 

Variables for Intelligent Tutoring.” Proceedings of the 2005 InSITE Conference, 2005, 

doi:10.28945/2890. 

Looi, Chee-Kit. Artificial Intelligence in Education: Supporting Learning through Intelligent 

and Socially Informed Technology. IOS, 2005. 

Rosé, Carolyn Penstein. Artificial Intelligence in Education: 19th International Conference, 

AIED 2018, London, UK, June 27-30, 2018: ProceedingsnPart 1. Springer, 2018. 

“THE VIRTUAL TUTOR - Development of a Virtual Tutor within a Simulated Working 

Environment.” Proceedings of the 4th International Conference on Computer Supported 

Education, 2012, doi:10.5220/0003966601910194. 

Weischedel, Ralph M., et al. “An Artificial Intelligence Approach to Language Instruction.” 

Artificial Intelligence, vol. 10, no. 3, 1978, pp. 225–240., doi:10.1016/s0004-3702(78)80015- 

0. 

Wenger, Etienne. Artificial Intelligence and Tutoring Systems: Computational and Cognitive 

Approaches to the Communication of Knowledge. Morgan Kaufmann Publishers, 1987. 


for Education 

SOMA S DHAVALA 

Premise and Problem 

Free and Compulsory Education for children in the age group of 

6-14 is a Fundamental Right, enacted by the Parliament of India 

[1]. It acknowledges, in spirit, that access to education is necessary 

to realising an individual’s inner potential. However, it is quite a 

feat to achieve even this relatively modest goal in countries such 

as India, where the varied demographic landscape renders “one 

size fits all solutions” ineffective [2]. In addition, the number of 

the aspirants, and the unfavourable skew on the supply side of the 

quality resources, makes it even harder. A larger supporting 

ecosystem is needed, wherein an individual is given 

attention based on his/her learning capabilities and the 

changing world order. Adequate quality learning resources at 

all levels (including physical, emotional, individual and societal) 

are also required. Can technology help? 

Prior 

For a long time, EdTech remained an academic activity, where 

both models of learning materials and methods, and the means 

of instructional delivery, such as those found in intelligent tutoring 

systems, were largely experimental and were without much 

penetration, despite strong theoretical foundations [3-4]. However, 

as the internet has become both accessible and affordable, 

Soma S Dhavala is an Innovator, Entrepreneur and AI Architect, operating at the interface of 

Statistics, Machine Learning, Computing and the Internet of Things. He is a co-founder of 

VitalTicks, a digital healthcare startup, - where he is leading product design, architecture and 

solutioning. Soma consults in the EdTech space, architecting AI solutions and platforms, and 

designs and reviews curriculum in AI and Deep Learning. 

Soma obtained his Ph.D (TAMU) in Statistics (where he worked on applying Bayesian 

Nonparametrics to systems biology) in 2010, Master’s (IIT-M) in 2000 and B.E (SRKREC, Andhra 

Pradesh) in 1997. He also worked in a post-doctoral role in the area of Dynamical Systems at 

TAMU for almost a year. He has over 10 publications and multiple patents to his credit. 

MOOCs and Flipped Classrooms have become successful, and 

access to quality content is no longer a barrier, at least in terms 

of higher education. A person in a remote village in India can 

enroll in a course taught and conducted by faculty at Stanford, 

which would have been unthinkable a few years ago. Despite the 

credibility of the offering and the relatively privileged learning 

groups, such platforms suffer heavy dropout rates, largely because 

either the intrinsic needs or the necessary prerequisites, or both, 

have not been met. As per the ASER 2017 report, unaddressed 

learning gaps typically carry forward, making the learner even 

more vulnerable to failure in future [6]. Given a deeper mobile 

penetration, and cheaper data access plans, can datahungry 

AI solutions deliver personalised education, and 

rival the breakthroughs happening in other disciplines 

such as Computer Vision and Natural Language 

Understanding (NLU)? 

Parts 

AAI-powered or an otherwise learner-centric EdTech 

offering needs to have, at the least, 1) presentable content and 

representation of its constituents 2) an assessment framework 3) 

a taxonomy of the concepts and lesson plans to chart learning 

journeys. Outside the realm of these formal pedagogic elements 

YOUTH VOICES 

ISSUE • 9 

4 7

YOUTH VOICES 

In order for AI-powered solutions to be 

contextualised, commoditised, and 

available to wider, diverse audience, a rich 

set of tooling, standards, and best practices 

are required. 

REFERENCES 

YOUTH VOICES 

lie the important but often overlooked elements such as 4) user 

experience, 5) behavioural and social contexts, and 6) benevolent 

marketing and branding. Without these conditions, chances of 

an EdTech solution being a successful product are rather slim. 

Nonetheless, there are efforts to push the envelope, and harness 

the seemingly ripe conditions for trying out AI powered solutions 

[7-10]. Any particular solution makes certain assumptions 

about those elements in the composition. Consequentially, the 

relevance and effectiveness vary for any learner. In order for AIpowered 

solutions to be contextualised, commoditised, 

and available to wider, diverse audience, a rich set of 

tooling, standards, and best practices are required. 

Product Building 

AI and Machine Learning are still evolving. Unlike physical 

products that can be perceived directly, such as an electric 

fan, AI products are most likely be intangible. For example, 

a manufacturer can provide a five-year warranty on the fan’s 

motor. A user can exercise the warranty when the motor burns 

out. Contrast that with an AI-powered content recommendation 

solution – what can be guaranteed or certified about it? This 

opens up a pandora’s box. Is there a way of building AI-powered 

products so that they conform to standards and scrutiny by 

third-parties? Are the results reproducible, auditable, scalable and 

customisable, among many other dimensions? The AI industry 

is responding to these challenges with conceptual 

frameworks [12], open standards and platforms [13-15], 

and AI education resources [16]. Another set of orthogonal 

concerns has far-reaching implications from a pedagogical 

and ethical standpoint. At a fundamental level, can the results 

be explained to the learner? Can the Right to Explanation be 

enforced, which could be enacted as a law in the future? Should 

AI-driven EdTech be regulated to safeguard learner interests? For 

example, a popular commercially available learning application is 

not in the purview of regulation. 

Conclusion 

The potential of personalised quality education, regardless of 

geographical, socio-cultural and other affiliations etc.., by means 

of AI and Deep Learning, is recognised [19-22]. However, a 

sustained, coordinated effort to address them holistically, from 

pedagogical to experiential to business to ethics and governance 

angles, is required. Concerted efforts are needed to develop 

a) data-first platforms b) shared NLU and AI infrastructure and 

c) agile processes to develop AI products. These products should 

eventually be used to create customised, relevant, engaging, 

explainable and governable solutions for every learner. 

[1] Right To Education Act, 2009 

[2] Still too many children out of school, The Hindu, September 4, 2018 

[3] Nikos, M., Hendrik, D., Katrien, V., and Erik, Duval., (2014), Recommender 

Systems for Learning, Springer-Verlag NewYork, 2014. 

[4] Advances in Intelligent Tutoring Systems, Eds. Editors: Nkambou, Roger, 

Mizoguchi, Riichiro, Bourdeau, Jacqueline (Eds.), Springer-Verlag Berlin 

Heidelberg, 2010. 

[5] Jordan, K. (2015). Massive open online course completion rates revisited: 

Assessment, length and attrition. The International Review of Research in Open 

and Distributed Learning 

[6] ASER 2017: Beyond Basics 

[7] ALEKS 

[8] ENLEARN 

[9] Sana Labs: AI for Education 

[10] expi.com 

[11] ReedEx: An Intelligent Reading Assistant 

[12] Fifteen Commandments and an Axiomatic Framework for building Machine 

Learning Products 

[13] openAI: Discovering and enacting the path to safe artificial general intelligence 

[14] AllenAI: AI for the common good 

[15] Wadhvani.ai: AI for social good 

[16] 500k.ai: 500 Thousand AI Programmers Initiative 

[17] Byju’s personalization engine 

[18] The EU General Data Protection Regulation 

[19] Dhar, V., Nilekani, N., S. Maruwada, Pappu, N. (2016), Big Data as an Enabler, 

Journal of Big Data, Vol. 4, No. 3, pp. 137-140 

[20] Kurshan, B, The Future of Artificial Intelligence in Education, in Forbes, 

accessed on 6th September, 2018. 

[21] China State Council’s “A Next Generation AI Development Plan”, 

[22] NITI Ayog’s discussion paper on “National Strategy for Artificial Intelligence 

document”, 2018 

All links were accessed on 6th September, 2018 

ISSUE • 9 

4 9

ACTIVITY 

BULLETIN 

Dr. Anantha Duraiappah elected as a Fellow 

to The World Academy of Sciences 

May, 2018 

Dr. Anantha K. Duraiappah, Director, UNESCO Mahatma 

Gandhi Institute of Education for Peace and Sustainable 

Development (MGIEP) was recently elected as a Fellow to 

The World Academy of Sciences (TWAS), a pioneering, 

merit-based academy focused on the advancement of science 

in developing countries. Dr. Duraiappah was recognised and 

elected to TWAS for his contribution to the progression of 

the field of sciences in developing nations.Speaking on his 

election, Dr. Duraiappah expressed: “It is an honour to be 

a part of this prestigious body of scientists who have carried 

out some ground-breaking work in science. I look forward 

to working alongside this body towards advancement of 

sciences – specifically in the field of learning sciences.” 

UNESCO MGIEP conducts a Plenary Session 

at the the 11 th Biennial Comparative 

Education Society of Asia (CESA) Conference 

May 11-12, 2018, Siem Reap, Cambodia 

The Comparative Education Society of Asia (CESA) is an 

association of educational scholars, working to promote 

comparative education as a scholarly field across Asia 

and to strengthen Asian voices in global debates about 

education. The 11th CESA Biennial Conference was held 

on 11–12 May 2018 in Siem Reap, Cambodia. The theme 

for 2018 was ‘Education and Social Progress: Insights from 

Comparative Perspectives’. On May 12, a plenary panel 

discussion was dedicated to UNESCO MGIEP’s seminal 

report Rethinking Schooling: The State of Education for 

Peace, Sustainable Development and Global Citizenship in 

Asia. Dr Yoko Mochizuki, Head of Rethinking Curricula 

Programme of UNESCO MGIEP, presented the key 

findings. 


The Government of Delhi organises a 

workshop for Special Educators in Screening 

Different Learners 

May 10, 2018, New Delhi, India 

In its mission to achieve inclusive and quality education for 

all (SDG 4.7), UNESCO MGIEP conducted a capacitybuilding 

workshop to enhance the skills of special educators 

in identifying and addressing students who learn differently 

as part of Government of Delhi’s project SMILE for its 

public schools. It was held on May 10, at Department of 

Education’s premises and attended by 60 psychologists 

from the National Capital Region. UNESCO MGIEP’s 

Difference Learning Project aims to improve functional 

literacy rate by enabling teachers, special educators, 

psychological experts and parents to identify different 

learners, provide diagnosis and cater to their different 

needs in the classroom. Dr Nandini C. Singh, Head of 

the Difference Learning Project at UNESCO MGIEP, 

conducted the workshop in collaboration with Ms Geet 

Oberoi, Founder of Orkids, an organisation working in 

the field of assessment, remediation and teacher training 

of educators.were discussed. The meeting also provided 

the participants with the opportunity to understand the 

work of their fellow partners for new collaborations and 

partnerships to emerge. 

Textbook Embedding Project going strong 

in Sikkim with the Second workshop for 

Textbook Authors 

May 20 – 22, 2018, Gangtok, India 

The project to embed concepts of peace and sustainable 

development in textbooks of Sikkim is progressing steadily 

with the second workshop for textbook authors organised in 

Gangtok from May 20-22, 2018. The embedding project is 

conceptualised as a continuous and comprehensive capacity 

building of textbook authors. The second workshop 

marked a beginning with textbook authors starting to create 

early drafts of examples of chapters, which went through 

participatory feedbacks as well as feedback from facilitators. 

ISSUE • 9 

5 1

ACTIVITY 

BULLETIN 

YESPeace Partners’ Meeting- India 

May 22-24, 2018, Gurugram, India 

UNESCO MGIEP hosted the 2018 Partners Meeting 

of the Youth for Education, Sustainability and Peace 

(YESPeace)Network at TERI Gram in Gurugram, India 

from 22-24 May 2018. The Partners’ Meeting brought 

together existing and potential partners of the YESPeace 

Network from across South Asia and Africa. The meeting 

served as a platform to develop a common and shared 

understanding of the YESPeace network, discuss the 

potential of cross-country and intra-country collaboration 

between YESPeace partners, identify timelines and orient 

the partners to MGIEP’s Youth Kindness Campaign. 

Different ideas and strategies were explored and plans 

for the growth and expansion of the network were 

discussed. The meeting also provided participants with the 

opportunity to understand the work of their fellow partners 

for new collaborations and partnerships to emerge. 

How should we prevent violent extremism? 

The youth may have the answer! 

May 25, 2018, New Delhi, India 

It may have been early on a Friday morning, but 

the 100 plus youth who attended the launch of 

#YouthWagingPeace—Action Guidelines for the Prevention 

of Violent Extremism (PVE) were raring to go with ideas, 

questions and tangible action points for PVE. The launch 

was followed by a riveting panel discussion on Youth 

Action Across Borders which saw panellists such as author 

and youth activist Gurmehar Kaur from India diving 

deep into their own experiences on youth empowerment. 

Underscoring the need for action-oriented guidelines, 

H.E Ms. Harinder Sidhu, High Commissioner to India, 

Australian High Commission in New Delhi said, “The 

#YouthWagingPeace Action Guidelines helps individuals 

and communities to build tools such as critical thinking, 

resilience, compassion and respect for diversity which they 

need to challenge extremist ideologies.” 


Three-day workshop for young educators on 

prevention of violent extremism 

May 22-24, 2018, Gurgaon, India 

The launch event of #YouthWagingPeace – Action 

Guidelines for Prevention of Violent Extremism was 

preceded by an intensive 3-day workshop for 25 young 

educators from across India, including three Rohingya 

youth working for the Rohingya refugee community in 

India. 

Facilitated by Ms Carolyn Nash, Coordinating Lead Author, 

#YouthWagingPeace and Mr Simon Kuany, Project Officer, 

#YouthWagingPeace, the 25 young educators were able to 

adapt the Action Guidelines and came up with their own 

context-specific lesson plans for their respective stakeholders 

(teachers, school administrators, families and policymakers). 

UNESCO MGIEP will continue to work with these young 

educators to ensure that the lesson plans are implemented 

and the outputs reflected/included in the broader 

#YouthWagingPeace workbook. This workbook will include 

all the tested lesson plans developed and tried by young 

participants of #YouthWagingPeace workshops. 

Learn to play and play to learn – a 

Distinguished Lecture hosted by UNESCO 

MGIEP on the role of gaming in education 

June 28, 2018, New Delhi, India 

Can learning be made fun and experiential? In UNESCO 

MGIEP’s Distinguished Lecture held at the UNESCO 

New Delhi, India on Thursday, June 28, 2018, Zigor 

Hernandorena Juarros, Senior Project Manager – Fun 

Learning Department, Ubisoft (makers of Assassins Creed 

Franchise, The Division, Far Cry, Watch Dog), discussed 

how digital pedagogies such as games can be used as an 

effective tool to impart learning. The lecture was attended 

by over a hundred participants including teachers, 

educators, curriculum designers, ed-tech experts, games 

designers and developers as well as students. In his lecture, 

titled ‘Learn to play and play to learn, the role of gaming 

in education’, Zigor explained how digital games can be 

designed to reward players, acknowledge practice and 

incremental progress, thus making learning more effective 

through peer-collaboration and a project-based approach. 

Zigor also underlined the need to develop suitable content 

for games to make them more relevant to 21 st century 

learners’ reaction from vision, and to turn the victim into 

the hero or the warrior for waging peace. 

ISSUE • 9 

5 3

ACTIVITY 

BULLETIN 

Youth for SDGs Peace Tables: Organised by 

Standing Together to Enable Peace 

June 29 - July 9, 2018, New Delhi, India 

Twenty youth leaders from New Delhi and other parts of 

India have created dynamic Community Action Plans for 

Art and Democracy, digital media, gender, and interfaith 

dialogue. The plans were developed through a series of 

Peace Tables from June to July 2018, organised by Standing 

Together to Enable Peace (STEP), a prominent Indian 

NGO working to build peace and a partner in UNESCO 

MGIEP’s Youth for Education, Sustainability and Peace 

(YESPeace) Network. Through the process of the Peace 

Tables, participants use their skills to combat the problems 

of incomplete and unequal implementation of the United 

Nations Sustainable Development Goals (SDGs) as well 

as the challenges to peace, prosperity and security in the 

context of a developing country like India. The Peace 

Tables successfully enabled the young leaders to dwell deep 

into their motivations, to sieve out reaction from vision, and 

to turn the victim into the hero or the warrior for waging 

peace. 

First ‘Training the Trainers Workshop’ on 

Cantor’s World 

July 24 - 25, 2018, New Delhi, India 

Second Futures Workshop on ‘The State 

of Social and Emotional Learning (SEL) 

Assessment’ 

July 19 - 21, 2018, Paris, France 

UNESCO MGIEP organised the Second Futures’ 

Workshop, held at La Sorbonne University, Paris between 

July 19-21, 2018, bringing together an expert group 

comprising of representatives of different disciplines, 

including neuroscience, human development psychology, 

anthropology, and education. The themes discussed 

were evidence-based practices for SEL implementation 

in education, and the commitment to bring nuanced 

cultural contextualisation to the implementation of SEL 

programming in schools. 


Compassion Acts, Empathy and Benefits 

of SEL – UNESCO MGIEP conducts an online 

presentation and discussion on SEL and 

Global Citizenship 

July 2018 

A key focus area for UNESCO MGIEP is content 

development in socio-emotional learning, as part of which 

the Institute delivered an online pre-recorded presentation 

on SEL as a key essence of achieving global citizenship. 

As part of the presentation, Dr. Nandini Chatterjee Singh, 

Programme Specialist –Science of Learning, posed some 

reflective questions to the educator community, including: 

1) Discuss one compassionate act you did today 2) Can 

you think of some ways to encourage empathy 3) Discuss 

how socio-emotional learning can benefit students’ overall 

learning. To view the presentation and read responses to the 

reflective questions, visit: https://bit.ly/2OvBpkc . 

UNESCO MGIEP’s Games for Learning Programme 

in collaboration with Fields of View conducted the first 

‘Training the Trainers’ Workshop on the game Cantor’s 

World. Cantor’s World, one of the many learning products 

developed by UNESCO MGIEP, is a simulation based 

game developed to understand the Inclusive Wealth Index, 

and to understand the uncertainties involved in governing a 

complex socio-economic system of a country. The game is 

targeted towards master’s level students of economics and 

sustainability studies and mid-career policy makers. The 

workshop was conducted as a first step of the pilot study 

to demonstrate the game to invited lecturers from different 

universities, introduce and train them on the game, in 

order to facilitate its deployment as a pedagogical tool in 

classrooms in these universities. The lecturers will go back 

and use this game as a learning resource in their classrooms 

and suggest improvements to the curriculum modules and 

the game itself. 

ISSUE • 9 

5 5

ACTIVITY 

BULLETIN 

Dr. Nandini Chatterjee Singh accorded The 

NASI – Reliance Industries Platinum Jubilee 

Award 2018 

Ecology, Economy and Society: Essays in 

Honour of Kanchan Chopra by Vikram Dayal 

(Editor), Anantha Duraiappah (Editor), 

Nandan Nawn (Editor) 

August 2018, India 

UNESCO MGIEP’s Director, Dr. Anantha Duraiappah 

recently co-edited a book titled ‘Ecology, Economy and 

Society: Essays in Honour of Kanchan Gupta’. Published 

by Springer, the book includes four vital themes related to 

the environment-development discourse -- sustainability of 

development, institutions and environmental governance, 

environment and well-being, and ecosystem and 

conservation. The book is a valuable resource for students, 

teachers, researchers, practitioners and policy makers. A 

Memorandum of Understanding (MoU) has been signed 

between Samsung India and UNESCO MGIEP to launch 

360-degree videos and VR educational content for 28 

UNESCO Heritage sites in India. 

#KindnessMatters: Indian youth show 

force of Kindness across the country on 

International Youth Day through Instameets 

in 14 cities on the Sustainable Development 

Goals 

August 12, 2018, India 

More often than not, youth are the subject of skepticism 

when it comes to stepping out of their comfort zones and 

taking tangible steps towards the Sustainable Development 

Goals (SDGs). However, on International Youth Day 

(Sunday, 12 August 2018) thousands of Indian youth 

display the power of kindness across 14 cities joined 

forces to perform voluntary Acts of Kindness towards the 

Sustainable Development Goals and proved naysayers 

wrong. The unique pan-India Instameets, organised by 

UNESCO MGIEP and Instagram India, were aimed at 

mobilising youth to take tangible action towards realising 

the Sustainable Development Goals, primarily in the 

form of collaborative acts of kindness. While these acts of 

kindness encompassed many issues, there was unanimous 

commitment by all youth groups across SDGs and cities. 

These unique Instameets marked the India launch of 

the International Youth Campaign on Kindness for the 

Sustainable Development Goals, which aims at collecting 

25,000 stories of kindness by youth across the world by the 

end of 2019. It is the first in the line of many international 

events celebrating youth action towards realising the SDGs. 


August 2018, India 

Dr. Nandini Chatterjee Singh, Programme Specialist – 

Science of Learning, UNESCO MGIEP was recently 

accorded The National Academy of Sciences, India 

- Reliance Industries Platinum Jubilee Award 2018 for 

Application Oriented Innovations covering both Physical 

and Biological Sciences. Dr. Singh is one of six recipients 

of this prestigious national award. At UNESCO MGIEP, 

Nandini oversees the Rethinking Learning Programme, 

including Libre, our flagship project and Difference 

Learning, amongst others. 

An Ahinsa Dialogue on ‘Creating a Culture of 

Peace’ – a conversation on peace, education 

and neuroplasticity 

October 2, 2018 | Paris, France 

In commemoration of the beginning of the year-long 

celebration of Mahatma Gandhi’s 150 th birth anniversary 

and the International Day of Non-Violence 2018, the 

UNESCO Mahatma Gandhi of Education for Peace and 

Sustainable Development (MGIEP) and the Permanent 

Delegation of India to UNESCO jointly organised the third 

Ahinsa Dialogue on October 2, 2018. The dialogue featured 

Sadhguru in discussion with Gregoire Borst, Professor of 

Developmental Psychology and Cognitive Neuroscience 

of Education, LaPsyDE, on ‘Creating a Culture of Peace’. 

Moderated by Dr Anantha Duraiappah, the discussion 

involved Sadhguru and Gregoire engaging in a conversation 

on ‘Peace, Neuroplasticity and Education’. Over 500 

participants attended the dialogue, including various 

dignitaries, educators, academics, policy-makers and youth. 

ISSUE • 9 

5 7

ACTIVITY 

BULLETIN 

#KindnessMatters: Youth across the 

globe come out in large numbers to launch 

UNESCO MGIEP’s International Youth 

campaign on Kindness for the SDGs 

October 2, 2018 | New Delhi, INDIA 

The global launch of UNESCO MGIEP’s campaign 

witnessed mass participation from youth in four countries--- 

South Africa, India, Pakistan and Mexico where UNESCO 

MGIEP’s International Youth Campaign on Kindness for 

the Sustainable Development Goals was simultaneously 

launched via a live link connecting the four nations. 

In true spirit of the campaign (which aims to mobilise youth 

acts of kindness towards the Sustainable Development 

Goals), the campaign launch witnessed massive acts of 

kindness committed by youth for one or more of the 

Sustainable Development Goals. 

#TAG e Kindness: Youth and policymakers 

dissect the role of “Kindness” in a time of 

Societal and Planetary change 

#KindnessMatters: Pakistan youth say ‘yes’ 

to Kindness; launch MGIEP’s International 

Youth Campaign on Kindness at YESPeace 

Festival 

October 2, 2018 | Islamabad, Pakistan 

#KindnessMatters: In a befitting celebration, 

South African youth join forces to launch 

UNESCO MGIEP’s Campaign on Kindness on 

the ocassion of International Day of Non- 

Violence 

October 2, Cape Town 

Africa Unite, in partnership with Activate! Change 

Drivers and the UNESCO MGIEP and Sustainable 

Development (UNESCO – MGIEP), Western Cape 

Provincial Government and City of Cape Town launched 

the International Youth Campaign on Kindness for the 

Sustainable Development Goals on October 2, 2018, at the 

Cape Town City Hall. 


October 2, 2018 | New Delhi, INDIA 

On the occasion of the United Nations International 

Day of Non-violence, and in line with its vision of 

mainstreaming the collective voice of youth in the highest 

levels of policymaking processes in education for peace, 

sustainability and global citizenship, UNESCO Mahatma 

Gandhi Institute of Education for Peace and Sustainable 

Development (MGIEP) hosted the latest edition of its 

flagship TAG e dialogue series as part of the Global Launch 

of the International Youth Campaign on Kindness for the 

SDGs in 4 cities: New Delhi, India; Islamabad, Pakistan; 

Cape Town, South Africa and TEC de Monterrey, Mexico. 

Over 130 young members of YESPeace Pakistan came 

together in Islamabad on October 2 to launch the 

International Youth Campaign on Kindness for the 

Sustainable Development Goals (SDGs). The launch took 

place at the Hill View Hotel wherein the youth shared 

their kindness stories with each other and video connected 

with the launch events in New Delhi, Cape Town and 

Monterrey (Mexico). The event was organised and managed 

by YESPeace Pakistan, a collective of youth organisations 

in Pakistan that included HIVE and College of Youth 

Activism and Development (CYAAD). The programme was 

supported actively by volunteers and facilitators from the 

International Federation of Red Crescent (IFRC) Pakistan. 

ISSUE • 9 

5 9

Learning with A.I. 

COMIC 

CORNER 

Seesaws are 

so much fun! 

I wonder how 

they work.. 

Oh look! 

There’s our teacher, 

we can ask her. 

Miss, when we get back 

to class, can you teach us 

how a seesaw works? 

Why later? 

We can learn that 

right now using 

Artifical Intelligence! 

See kids, A.I. can 

make learning fun 

and possible anywhere!

The Blue DOT is UNESCO MGIEP’s bi-annual publication, featuring articles showcasing our activities 

and areas of interest. The magazine’s overarching theme is the relationship between education, peace, 

sustainable development and global citizenship. To view the e-publication, visit – 

http://mgiep.unesco.org/the-bluedot

The Blue Dot Issue 9: Artificial Intelligence and the Future of Education

Create successful ePaper yourself

Delete template?

Save as template?