June 16, 2020 xerxes

In this interview De Kai talks about Artificial Intelligence from his cross-disciplinary perspective. We talk about the meaning of creativity, consciousness and mindfullness for modern AI, various misconceptions about modern AI and whether it is still programmable in the classical sense at all, about AI ethics and what AI means for us as a society, humanity and our future.

About De Kai

De Kai is Professor of Computer Science and Engineering at HKUST and Distinguished Research Scholar at Berkeley’s International Computer Science Institute. He is among only 17 scientists named Founding Fellow by the Association for Computational Linguistics, for his pioneering contributions to machine learning foundations of machine translation that led to the Google/Microsoft/Yahoo/etc translators. De Kai was recruited as founding faculty of the University of Science and Technology in Hong Kong directly from UC Berkeley, where his PhD thesis was one of the first to argue for the paradigm shift toward machine learning based natural language processing. He holds a Kellogg-HKUST Executive MBA and a BS in Computer Engineering (Phi Beta Kappa, cum laude, Revelle College honors) from UCSD.

For his work on AI, machine learning, machine translation, natural language processing, music technology, computational creativity, Debrett’s HK 100 recognized De Kai as one of the 100 most influential figures of Hong Kong.

For his work on AI ethics and society, De Kai was one of eight inaugural members selected by Google in 2019 for its AI ethics council.

See http://dek.ai for details.

De Kai’s Links

Website: http://dek.ai

Universal Masking Project: http://dek.ai/masks4all

Twitter: @dekai123

Facebook: https://www.facebook.com/dekai1

YouTube, 26 Apr 22, Visual simulations show why we all need to wear masks now https://youtu.be/yfeW2l8G_W4

South China Morning Post, 25 May 2020, Coronavirus spread would dramatically drop if 80% of a population wore masks, AI researcher says https://www.scmp.com/video/coronavirus/3085971/coronavirus-spread-would-dramatically-drop-if-80-population-wore-masks-ai

Boma COVID-19 Summit, 23 Mar 2020, The disastrous consequences of information disorder erupting around COVID-19: AI is preying upon our unconscious cognitive biases https://youtu.be/ZidC7oRd7Pc

TEDxChiangMai, Thailand, 7 Sep 2019, The Paradox of AI Ethics: Why Rule-based AI Ethics Will Fail https://www.youtube.com/watch?v=rKPhvb_9taw

TEDxOakland, California, 18 Nov 2018, Why AI is impossible without mindfulness https://www.youtube.com/watch?v=w_Sd2ZPPhv8

TEDxKlagenfurt, Austria, 16 Jun 2018, Artificial Gossips https://www.youtube.com/watch?v=hHyjSgCoNlw

TEDxBlackRockCity, Nevada, 29 Aug 2017, Artificial Children (no video due to technical difficulties)

TEDxZhujiangNewTown, Guangzhou, 14 Jan 2017 Why Meaningful AI is Musical https://v.youku.com/v_show/id_XMzA4MDQwODE4MA==.html or https://v.qq.com/x/page/i05088lu78h.html

TEDxBlackRockCity, Nevada, 31 Aug 2016 (no video due to technical diff

TEDxXi’an, China, Jun 2016. Surprise! You already have kids and they’re AIs https://v.youku.com/v_show/id_XMTc2MTI3NTc1Ng==.html

TEDxBeijing, China, Jan 2016, Can an A.I. Really Relate? What’s Universal in Language and Music https://v.youku.com/v_show/id_XMTcxMzYxMjEwNA==.html?refer=seo_operation.liuxiao.liux_00003307_3000_z2iuq2_19042900 or https://v.qq.com/x/cover/tuazku0z8e6fw2h/r0326z8r7ta.html

TEDxWanChai, Hong Kong, Aug 2014, Do You Speak Pentatonic? The Multilinguality of Music https://www.youtube.com/watch?v=zMAU8SY8pMs

TEDxElsaHighSchool, Hong Kong, Apr 2014, Music in Translation https://www.youtube.com/watch?v=eIFBNBGSJnk

TEDxHKUST, Hong Kong, Mar 2012, ReOrientate (no video due to technical difficulties)


Video Excerpts From the Podcast

Meet De Kai

Role of Creativity in Machine Learning and AI

AI Ethics, Society & Paradigm Shifts​

Artificial Intelligence and Coronavirus Prevention​

Human Level Artificial Intelligence​

Artificial Intelligence and Rethinking the Principles of Society

Transcript of the Interview
This text has been auto-transcripted. Please excuse mistakes.

Xerxes Voshmgir: Welcome to Challenging #ParadigmX.

My name is Xerxes Voshmgir.

And in my podcast interview people who challenge the status quo.

Can we have artificial intelligence without consciousness, creativity, or mindfulness, uh, language and music.

The only things that humans do better than any other species.

And what would that mean for the development of artificial intelligence?

Can we program ethical rules into the AIS or can we only teach them like we teach children?

And if so, what would that mean for us as societies as humanity?

Do we need to rethink the founding principles of society in an AI era?

Xerxes Voshmgir: Today, my guest is De Kai.

De Kai is one of the world’s leading experts in the field of artificial intelligence and one of the early pioneers of machine learning.

And that paradigm shift in AI.

I’ve always work on AI machine learning machine translation, as well as natural language processing, music technology, and computational creativity.

He was recognized as one of the 100 top influential figures of Hong Kong, where he is also professor of computer science and engineering at the Hong Kong university of science and technology.


He is a distinguished research scholar at Berkeley is international computer science Institute.

Apart from that, he is a musician and is involved in the field of social impact to enhance the dialogue between different cultures.

This is also reflected in his work.

He used his intercultural skills and knowledge, as well as his deep understanding of music and creativity to develop the world’s first AI for web translation that led to the Google Yahoo and Microsoft translators.

And for his work on AI ethics and society, the Chi was one of eight inaugural members selected by Google in 2019 for its AI ethics council.

Xerxes Voshmgir: so if you’re interested in artificial intelligence from the cars, cross disciplinary and paradigm shifting perspective, stay tuned.

Xerxes Voshmgir: Hi, my name is and today I’m here with De Kai.

It’s good to have you here De Kai.

De Kai: Great to be here.

Thanks for having me.

Xerxes Voshmgir: Yeah.

Thank you.

So, yes.

Could you please introduce yourself?

Who are you?

What do you do?

De Kai: Well, so, De Kai: I’m on the one hand, I’m an AI professor.

De Kai: if you know, things like Google translate or Yahoo translate or Microsoft translate, I, De Kai: pioneered those sorts of things.

I built the first web translation, AI and developed a bunch of the machine learning foundations for them.

De Kai: I also work on, De Kai: music technology, De Kai: AI for computer music and, De Kai: computational social science, De Kai: is one of the, De Kai: people who sort of, De Kai: you know, pushed machine learning approaches into the field of AI at a time.

De Kai: 30 some years ago when the field was dominated by what we call good old fashioned to AI, De Kai: which is will-based, you know, manually coded logic based, De Kai: systems that, De Kai: were really impossible to scale up.

De Kai: we, we, De Kai: fought this battle to like move away from logic, De Kai: stop trying to manually

model the world by writing everything we knew about the world by hand and instead model the mind, De Kai: which is not a logic machine, De Kai: you know, model these massively parallel distributed units that fire different activation levels at each other.

And, De Kai: from that complex system emerges a learning behavior.

De Kai: and so.

De Kai: you know, I’ve, I’ve been pushing on that the whole time.

I pretty much focused on things that have to do with getting cultures to relate better to each other, to understand each other better.

De Kai: That extends also to my work as a musician, De Kai: having formed one of Hong Kong’s, De Kai: world music groups.

De Kai: As well as, De Kai: my social impact work, De Kai: where, De Kai: you know,Afighting discrimination and, De Kai: looking to enhance dialogues between different cultures, De Kai: and, and looking at ways to decrease the, De Kai: alarming amount of polarization, De Kai: divisiveness, fear, and hatred.

De Kai: that we’re seeing both in, in many countries domestically, as well as globally in the geopolitical situation.

Teil 2

Xerxes Voshmgir: Okay, great.

And, Xerxes Voshmgir: please tell us why do you do what you do?

De Kai: No, I think I’ve had the privilege of, of being shown a lot of the world since I was a child.

My, my parents, De Kai: Took us all over the U S as you know, they, they were, De Kai: they had come to the U S as teenagers from war torn, De Kai: China and Hong Kong and Japan, De Kai: and so on.

And they, De Kai: landed in the mid, literally smack in the middle of the country in St.

Louis, you know, which is, De Kai: as you know, it’s, you know, on the Mississippi river, which is the separates the East and the West in the U S and all the Mason Dixon line, which separates the North and the South.

So literally right in the middle.

And I grew up in the Midwest, De Kai: heavily and in Illinois, De Kai: in the next state over, De Kai: it was, De Kai: very mind opening to see.

De Kai: the cultural differences, how they weighed on perceptions of people.

And then, and then my parents took us, De Kai: not only to almost every state in the union, but to, De Kai: other countries.

And, you know, we saw things about the developing world, De Kai: that were very, very different.

De Kai: we saw extreme poverty.

De Kai: we saw.

Things that made me question from an early age, whether all of the assumptions, De Kai: that, De Kai: were mainstream around where I was growing up, De Kai: maybe were relative to the, De Kai: environment, the, you know, the, the economic level, the education level.

And of course the cultural background that you’d have to deal with it.

So I think that these are things that.

As we’ve seen in the last decades have Le led to enormous stresses in, in humanity.

And, De Kai: that has really driven me, I think, to look for ways that between the technology and, you know, the arts music, De Kai: and just, De Kai: dialogue, social dialogue, De Kai: that.

There are things that are urgent.

De Kai: upon those of us who are able to, to look at how we can enhance how we give them really ramped that up, De Kai: so that we don’t end up with humanity.

He literally tearing itself apart and destroying itself in the planet, in the process.

Xerxes Voshmgir: so I’d be very interested to see your Xerxes Voshmgir: perception or perspective on like, what influence does music have in your life?

And this specifically, do you see a connection to your work in artificial intelligence and how it lets you perceive artificial intelligence?

De Kai: Oh, very definitely.

De Kai: you know what, the thing about intelligence human intelligence is that.

Language and music are basically the only things that we humans do better than, than other species.


I mean, we don’t, we don’t run faster.

We don’t fly.

We’re not very good at climbing trees anymore.

De Kai: we’re not stronger.

We don’t see more sharply.

We don’t smell more sharply.

We’re pretty lame as a species except for our unique abilities in language and music.

And the thing is that.

You know, De Kai: really music was the foundation of a lot of human language abilities.

Our, our ancestors were singing before they were talking and, De Kai: you know, early speech probably evolved from, De Kai: using our song abilities, De Kai: to be able to represent ideas about the world and.

Once we evolve stronger linguistic abilities, then we turned around and we applied those to music.

And so we’ve been developing more sophisticated types of music using musical languages.

De Kai: so this has like co-evolved through, you know, a million years and it leaves us with, De Kai: brains that are supremely evolved for music and language and, and those fundamental building blocks underneath it is how we implement intelligence and therefore studying.

Music is heavily related to studying language is, you know, the same areas of the brain that are used, except that music is activating.

Many of them in parallel, it’s like exercising those linguistic abilities on steroids, De Kai: and, De Kai: the kind of creativity, the kind of improvisation and expression in traditional music.

Teaches us a lot about how the mind creates, how, not just, how does the mind interpret new things that it’s never experienced before, but also how it learns and then is able to generalize.

So that again, take what it’s learned previously, but then express new things in real time.

De Kai: not in the sense of this, you know, very recent.

De Kai: aspect of, of music, especially in Western European symphonic tradition, where, you know, some, somebody writes out a complex score and then you have trained musicians that just do nothing, but they read an execute that score.

De Kai: that’s actually a very artificial recent form of music.

De Kai: that only happened in the last few hundred years.

Traditionally music is, is improvisational.

De Kai: and.

De Kai: it has conventions, but it’s not the same twice in a row.

De Kai: and you have a conversation between the musicians who are typically sitting there bouncing off of each other, just like you and I are talking.

De Kai: that’s that tells us a lot about how the brain works about, De Kai: what kinds of facilities there are to, with which to learn and to express and to create.

Xerxes Voshmgir: And, Xerxes Voshmgir: I have the thought now why we’re talking, maybe it’s a fall off.

Xerxes Voshmgir: but do you think that artificial intelligence, the way it has been created the last, Xerxes Voshmgir: couple of decades up to.

The time where you will know better than me when the point was, if it’s at all the case, deep learning was introduced that there was maybe a switch when it compared to music to the Western type of music, the artificial intelligence before, and got has somehow through the work of certain people, including you somehow to this more linguistic improvise creative type of approach, how to program artificial intelligence.

Xerxes Voshmgir: Was there such a change at all?

When you look at the deep learning, for example, or other stages in the development of artificial intelligence?

De Kai: Yeah.

Yes and no.

I mean, De Kai: I think the major change was, De Kai: what I was referring to earlier, when we move from rule-based logic based systems that were coded by hand.

De Kai: so that all the knowledge of the system was written by humans rather than learned by the machine.

De Kai: and then subsequently the machine learning models, De Kai: the, in terms of the deep learning models, De Kai: You know, there are one family of machine learning models.

And, De Kai: to be honest, I started playing with what we now call deep learning in the late 1980s, early 1990s.

In fact, my, my postdoctoral year in 1992 at university of Toronto, I spent happily playing, De Kai: in Geoffrey Hinton slash Jeff, De Kai: is the granddaddy of deep learning.

And I was literally working on what we now call deep learning.

De Kai: models at the time.

De Kai: but there are also many other machine learning approaches.

Deep learning is not the.

You know, be all end all.

De Kai: even Geoff Hinton himself has said on the record, I think we need to throw it all away and start from scratch.

De Kai: and, De Kai: you know, it’s just, it’s a phase, so there’s a lot of media hype.

De Kai: it’s fairly shallow about deep learning these days.

Those of us who are really working in AI research have already moved on to, De Kai: you know, that’s one, one of the tools in our toolkit, but like, De Kai: we’re looking at deeper problems, De Kai: like the, De Kai: issues that I’ve spoken about, De Kai: on, De Kai: modeling actually, De Kai: awareness, De Kai: mindful AIS.

De Kai: you can’t really have an AI that is just mindlessly doing deep learning.

De Kai: all that is is, De Kai: glorified regression.

De Kai: and, De Kai: it’s not aware of what’s it’s doing, De Kai: that’s a very important distinction between human level intelligence and, De Kai: sort of instinctive automatic, De Kai: reflexive types of learning that other species also have.

De Kai: so I think when we look at music, De Kai: you get the same dichotomies in music.

So, you know, for those of us who grew up playing musical instruments, De Kai: by.

You know, I, I don’t even remember not playing my, my, my family tells me that I was picking out melodies when I was two years old on the little kiddos.


And so, Nope, nobody had taught me to read music.

Nobody had taught me technique.

This was just something that you learn by, by the sort of associative pattern recognition, the same way kids learn language.

Nobody says to a kid, Hey, you’re not allowed to speak until I’ve taught you the rules of grammar and how to read.

And, and then.

Once I’ve done that, then you’re only allowed to say things that are when you’re reading from the page written by some, De Kai: poet.

De Kai: you know, it’s, it’s done automatically below the level of consciousness as you’re, as you’re hearing.

Someone speak.

You’re not saying, Oh, hearing that was a verb.

De Kai: someone that was the, De Kai: agent that was speaking, you’re not doing that.

It’s all happening at the unconscious level.

De Kai: and so it’s very much the same thing with music is that if you’ve learned music in this natural way, as opposed to the recent artificial way, De Kai: then that teaches us a lot about how the, how the mind works.

But at the same time, once you’ve done that and you start creating and you start thinking, how would I compose a piece?

You know, or if I’m a DJ, how would I like construct this track?


De Kai: Now you’re applying the linguistic side now you’re, De Kai: like really thinking that way that a poet would think.

De Kai: and so you’ve got all those different types of mental processes going on and studying music helps us also to understand that about the nature of human intelligence.

Xerxes Voshmgir: Okay.

And the way artificial intelligence is programmed today, and the research it’s done today goes also more into this direction.

De Kai: I think it’s going to increasingly, De Kai: it is not very much out in the, De Kai: popular, you know, De Kai: awareness, popular press, De Kai: yet, but, De Kai: people who are deep in the, in, in the forefront of AI research are, are thinking about these problems that are making proposals.

You know, our, our lab is certainly integrating, De Kai: Both the sorts of automatic unconscious, you know, fast, fast pattern recognition and associative types of processing with more mindful styles of, of, De Kai: consciously control understanding of what it is I’m doing.


Xerxes Voshmgir: And how can we imagine this?

I mean, what does it mean consciously controlled, for example?

De Kai: well, so, you know, there are psychologists, De Kai: for many decades now, De Kai: have been studying the differences between automatic mental processes and controlled or conscious mental processes.

So, you know, De Kai: Daniel Kahneman, De Kai: who wrote the bestseller thinking fast and slow, De Kai: it calls it system one versus system two.

De Kai: and so, De Kai: there’s a, De Kai: an enormous amount of empirical data on the characteristics of system one or automatic mental processing versus system two or controlled, De Kai: conscious, De Kai: mental processing where you’re aware of what you’re thinking, you know, like, so, so the thing is that if you’re able to introspect, if you’re able to describe your thoughts.

That is conscious controlled reasoning.

De Kai: But it, you know, something like the fact that you’re not thinking, Oh, hearing was a verb, right.

That that’s not conscious, that’s unconscious processing.

De Kai: or, you know, if I ask you for example, De Kai: like D do you know what the difference is between walking and running?

Xerxes Voshmgir: I think so well walking.

Well, I mean, the way you ask it, I wonder if I’m perhaps all no, but walking is.

A slow type of movements compared to running, which is a fast type of movement.

De Kai: Sure that’s true.

But, but you also, like, you’ve probably seen some of these, De Kai: competitive speed walking, De Kai: races, right?

So they’re walking very fast.


And you can run slower than you can

Xerxes Voshmgir: yeah, yeah,

De Kai: there is actually a technical difference.

Now let me just make sure.

You do know how to walk.


And you do know how to run, right?


Xerxes Voshmgir: Yes.


De Kai: And yet you don’t know what the difference is.

The difference is that if you’re running both feet, leave the ground at the same instant.

Now, you know, most of us know how to walk.

Most of us know how to run and most of us have no idea that that is the difference.

That is unconscious mental processes.

But now that we’ve talked about this now you’ll have conscious knowledge of it.

Xerxes Voshmgir: Okay.

And does it mean there’s a way to program artificial intelligence and the way that it becomes conscious about the processes

De Kai: Yes.


Xerxes Voshmgir: and the old.

De Kai: yeah.

That is exactly what needs to happen.

Or AI to actually reach human level.

Xerxes Voshmgir: Okay.

Well, I mean, that also believe, but do you believe it’s possible because some people say it’s not possible.

De Kai: Well, I think people get confused about what it means, right?

Because as soon as you drop in a word like consciousness, De Kai: I mean, you know, there’s like nine different definitions of consciousness.

I, and, and, De Kai: they all depend on, on assumptions, right.

That there’s a.

Some small group of core axioms that you have to take on what you mean.

Like, first of all, De Kai: you know, De Kai: are we talking about mind, body dualism?


So, De Kai: are we looking at, De Kai: issues of the soul, De Kai: or, De Kai: you know, which, which a lot of people fold into what they are casually thinking consciousness meets.

But, De Kai: in the sense that I’m talking about this, De Kai: it is the consciousness that I’m speaking of is empirically measurable.

De Kai: so we’re not attempting to say things about, De Kai: the metaphysical assault.

De Kai: but we’re, De Kai: looking at things that, De Kai: you can actually read from brain scanners from MRI or EEG.

De Kai: we’re looking at speed, De Kai: issues, right?

So there’s a lot of experimental psychology, De Kai: cognitive psychology that you can do with everything from eye tracking to, you know, just how fast people can process things, what kinds of mistakes they make.

De Kai: right.

And so there’s a lot of empirical stuff that you’ve been observed about unconscious versus conscious mental processes.

Xerxes Voshmgir: Okay.

All right.

And, Xerxes Voshmgir: how do you see the connection between consciousness and intelligence?

And maybe before you tell me that, could you tell me what your definition of intelligence is?

De Kai: Well, so, De Kai: there are different approaches toward defining intelligence and, De Kai: you know, one of them in the style of Turing of course, is more of a behavioral definition.

So if it.

If it looks like it looks like a dog and it smells like a dog and walks like a dog.

It probably is a dog.

De Kai: so that’s basically the style of definition, Turing had to in his famous Turing test, De Kai: I think from a science point of view, that is a reasonable definition.

De Kai: if you can drop an AI into a new situation that has never experienced.

If it’s had the same kinds of environmental upbringing as a human, and if it reacts the same way as a human, then you’ve produced the human level AI.

De Kai: and again, De Kai: not ascribing any, De Kai: qualities of metaphysical qualities of having a soul to that definition.

De Kai: but that would have achieved human level AI.

And beyond that, if it becomes more powerful in its learning capabilities or its reaction speed, then it becomes, starts becoming a super intelligence.

Xerxes Voshmgir: Okay.

And, Xerxes Voshmgir: and the context of creativity, do you see one there as well?

A connection between intelligence, consciousness and creativity.

De Kai: Oh, absolutely.

I mean, I think creativity is a.

It is, it is basic.

So, so there are a couple of very common cliches that you hear as soon as people start talking about AI casually, which is that, Oh yeah, you know, these machines are getting incredible, but, De Kai: they’ll never have be able to, De Kai: have human qualities like creativity and emotion and, De Kai: those sorts of things.

And, and, and the thing is, I think that that’s actually misleading.

De Kai: so when it comes to, De Kai: well, when it comes to emotion De Kai: That is actually simpler, more primitive and easier to model, then, then, you know, truly intelligent process, human, intelligent processes, like language, understanding anybody who’s had a dog knows that emotion is very basic.

You don’t need human level intelligence to have emotion.

De Kai: It’s far easier to model that it’s just, you know, people didn’t start doing it until recently, but it’s easy to do compared with a lot of other things.

De Kai: and as for creativity, De Kai: it is pretty much impossible to have any kind of intelligence without creativity.

You cannot have a learning system without creativity.

By definition.


So, so like, what does it mean to learn?

It doesn’t mean to memorize memorization requires no creativity.

That’s true.

But to actually be intelligently learning is not just a memorize, like a database does it’s to have been exposed to a few examples, but then you leap to generalizations and you are making inferences in situations you’ve never encountered before.

That by definition is a creative process, to be able to speak a sentence that in a new situation like we’re doing now, that we’ve never done before is requiring creativity.

De Kai: it’s, De Kai: these are fundamental building blocks of machine learning and you cannot have, De Kai: intelligence without those creative building blocks.

So I think, you know, It’s a comforting myth for a lot of people that, Oh, machines can’t replace humans because we’re uniquely creative and, you know, emotion.

And all of these things are, you know, cannot be replaced.

It’s, it’s a sentimental notion that doesn’t have any scientific basis.

And so I understand very much the human yearning for that, but I think we need to redirect that human yearning.

To how do we with, De Kai: how, how do we with, with, De Kai: positivity with, De Kai: you know, human, the best of the human qualities for creating, De Kai: empathetic societies deal with the fact that.

These new machines are already creative and are already very much, especially in social media, reflecting emotion, De Kai: how, you know, those things are altering our societies in drastic ways.

If we look at the kinds of polarization that I was talking about at the outset, The reason that so much of my work in the last several years has been directed in AI ethics and AI in society is because of the fact that AI powered social media and media and search engines are ripping apart our societies as an unintended consequence of the fact that they’re actually amplifying.

De Kai: all these natural human, De Kai: evolutionarily hardwired qualities, De Kai: that cause us to be biased, to believe things that we probably shouldn’t believe, De Kai: and to react in ways that are irrational because in the old days, your tribe

would be, you know, Killed by another tribe, if you didn’t react that way, but in a modern era where weapons of mass destruction are being democratized so that like ordinary people can be carrying really, really dangerous weapons of mass destruction.

It’s no longer tenable is no longer survivable.

For society to be that way.


And so we need to figure out how to redirect those sentiments about human emotions, about taking care of each other and love and, and, De Kai: understand how in our new era where society is heavily populated by AI’s already today.

How do we bring those qualities to bear in that mixed society of humans and AI, rather than allowing fear and hatred and stronger emotions to be learned and propagated by the AI.

Xerxes Voshmgir: So Xerxes Voshmgir: in this context, I wonder what your, Xerxes Voshmgir: perception or perspective actually is when it comes to.

The term artificial, where we talk about intelligence.

And I mean that in the way that I have the impression that when we call artificial intelligence, artificial, Xerxes Voshmgir: of course it means it’s not natural, but at the same time,

Xerxes Voshmgir: You could also get the impression that it is not, Xerxes Voshmgir: equal or is not, is not.

Xerxes Voshmgir: because it’s just artificial.

It’s not real

De Kai: Okay.

Xerxes Voshmgir: this could be an interpretation and this is also when I talk to people and hear people talk about artificial intelligence.

This is also something, it seems to me that gives them some safety because it’s not real.

So it cannot be also a real threat.

I don’t necessarily believe that artificial intelligence is the ultimate threat or something, but I believe it’s important how we go forth, the way we program it.

So what what’s, what’s your take when, when it comes to the term artificial, do you think it is the proper term?

When we talk about this type of intelligence?

De Kai: Well, sure.

I actually do.

And here’s why, De Kai: You are an artificial intelligence.

De Kai: well, look at the Oxford definition of artificial.

De Kai: I.

Pretty close to word for word, De Kai: made or created by human beings, roughly as a copy of something occurring naturally.

So pretty much I think we can agree that you would not exist unless you were made by human beings and that you were made rough as a rough copy of something that occurred naturally.

De Kai: And since obviously you’re intelligent by definition, you’re an artificial intelligence and we all are so, De Kai: I think it’s actually a very appropriate term.

Xerxes Voshmgir: Okay, but, but this is a perspective that most people don’t.

Take as you do, of course, because most people don’t consider themselves as artificial intelligences.

De Kai: Yeah.

So I’m

Xerxes Voshmgir: this is actually what’s that.

De Kai: I’m working to

Xerxes Voshmgir: Okay.



I see.


All right.

Yeah, but I mean, it, it, you could then also say, Xerxes Voshmgir: if you turn it around that artificial intelligence, Xerxes Voshmgir: computer based artificial intelligence, Might be in a way similar to human intelligence or what we call natural intelligence.

De Kai: Yeah, indeed.

I think it, De Kai: we’re getting closer and closer to.

De Kai: models in the field of AI that are getting more natural.

We’re still quite a distance off.

De Kai: but, De Kai: I think that for AI to progress, it has to get more natural.

We can, you know, AI is, can not be mindless.

They’ll never be intelligent that way.

De Kai: today’s AI is because they’re mindless.

They require exponential amounts of data.


And so, De Kai: people have, have confused AI with big data.

If you’re going to process big data, you also need, you know, exponential amounts of computation power, which is why you have these enormous farms of CPUs and GPUs these days.


But when we look at a human brain, you know, it’s being powered by.

Less electricity than a 100 Watts electric light bulb.

You don’t need that much computation or that much data.

If you have human level intelligence, Hey, a true AI is actually small data to be able to make those intelligent generalizations for the same amount of small data that a three year old kid learns from.

So there it is inevitable that progress in AI will drive us closer to natural intelligence.


Xerxes Voshmgir: Okay.

And is there a timeframe that you personally believe is realistic?

De Kai: I am always reluctant to say this, because one of the terrible things about the field of AI is that it has a long history of always predicting that the problem would be solved in 20 years.

And so if I say the problem will be solved in 20 years, 20 years later, people will still be saying that that said, De Kai: progress has been, De Kai: incredible, De Kai: you know, going back from.

Say about 30 years ago when I was finishing my PhD and still fighting the battle against good old fashioned AI, which was, you know, hand coded logic rules.

We have made enormous progress in what AI can do today, you know, for the people doc, then what we are doing today would be mind blowing.

I have no doubt that that will continue to be the case.

Xerxes Voshmgir: Okay, Xerxes Voshmgir: so maybe let me rephrase.

Xerxes Voshmgir: do you believe it will be in less than 200 years?

De Kai: Yes,

Xerxes Voshmgir: so I’ve got another question I’m really interested in, Xerxes Voshmgir: Your work on the coronavirus prevention was all over Xerxes Voshmgir: the media, even, Xerxes Voshmgir: president Obama retreated an article where your statements when they’re, Xerxes Voshmgir: please, Xerxes Voshmgir: give us some backgrounds.

De Kai: Oh, so this was something that happened, you know, like, like many people, obviously we’ve been really struggling with, how do we, how do we deal with, De Kai: the horrible effects of the coronavirus and, De Kai: because.

I have joint appointments at the Hong Kong university of science and technology.

And at Berkeley, De Kai: I see, you know, very regularly what’s going on, both in the East and West.

De Kai: and, De Kai: by February it was really alarming me.

That Hong Kong had, was paying attention to the who warnings and the warnings from China.

And it was already locking down in January already.

De Kai: there was immense amounts of funding for what happens on February 1st when the semester was supposed to begin.

De Kai: should we not physically open up?

And, and indeed we’d never opened up physically.

We, we converted to all online teaching the entire semester.

De Kai: And in by mid February.

De Kai: it was really alarming to me that not only was that discussion, not really happening seriously in the West, in the us or Europe.

De Kai: but, De Kai: even, you know, face mask wearing was not being discussed.

De Kai: Hong Kong already, you know, literally 99% of the population massed up.

If I could not go into town, not wearing a mask without hearing about it.

And getting dirty looks and, you know, De Kai: because, De Kai: you’re not doing your part to prevent spreading your possible Covid viruses to other people around you.

De Kai: it was not just about protecting yourself.

It was about protecting those around you, that community around you.

But when you look at the WHO data, when you looked at, you know, other datasets, Johns Hopkins or whatever, even though they were listing, which countries had social distancing and which countries had testing and contact tracing, they didn’t even bother listing which countries had masking.

And this was alarming to me because masking costs far less then testing and tracing and many other measures, there is no more evidence for social distancing than there is for masking.

And, De Kai: and so we formed a team rapidly in, De Kai: late February, De Kai: with, De Kai: not only a me at a Hong Kong at Berkeley, but also with, De Kai: London, De Kai: and Paris and Helsinki and Estonia Tallin, De Kai: across disciplines from everything from medical doctor in,

De Kai: London on the front lines of treating COVID, De Kai: to, De Kai: economist in Paris, De Kai: to, De Kai: someone who is actually in a senior advisor to the ministry in Finland, De Kai: looking at public policy, De Kai: and a computational biologist, De Kai: out of Cambridge.

And so we, we, we really looking broadly at this, we built two new theoretical models.

De Kai: I built an AI inspired model, De Kai: That was combining agent based modeling from the long tradition in AI of, De Kai: you know, you have intelligent agents walking about and interacting when they meet each other, De Kai: to, De Kai: do the predictive modeling.

And, De Kai: I made it highly visual.

So that you could actually see these agents walking around and you can just take your browser and go straight there and play with it yourself.

If you like, and you can set your own assumptions about how many people are infected and how many people there are and how effective your masks are at blocking transmission of viruses versus blocking absorption from other people into your system, when you inhale and you can see what happens.

And so I made this four minute video to explain as a tutorial, how to use that.

De Kai: the video got a huge amount of interest, De Kai: and, De Kai: led to, De Kai: you know, a quote cover story on Vanity Fair.

De Kai: that really raised a lot of awareness, De Kai: amongst European and, De Kai: North American States.

And, De Kai: basically we found, De Kai: that if you need 80 or 90% of the population, To wear face masks and you need them to do it fast before the virus spreads to too much of the population.

If only half the population works mass, it has almost zero impact.

And the reason the visual simulation is so good is because that’s counter intuitive for most people.

Most people think if, if half the population are wearing masks, then it’ll stop half the infections.

And that’s simply not true in the turns out because of the exponential behavior of virus spread.

De Kai: half the population wearing knots has almost zero effect, De Kai: 80 to 90% is needed in order to have significant impact.

And if you do that and you combine it with other sensible measures, social distancing, De Kai: we do need testing and testing.

Then you can control the spread.

Long enough for the scientists to develop the vaccines and the treatments and prevent the hospitals from getting overloaded.

Otherwise we create thousands of unnecessary deaths a day.

De Kai: and so, De Kai: that project was a really cool application of various concepts from AI and computer science to, De Kai: tackle the, the social problem right now.

That is urgent upon us that we’re still grappling with.

Of how to say it as many lives as possible, how to be able to exit, De Kai: from extreme lockdowns, De Kai: without risking lives in an excessive way in Hong Kong, because we put on masks really early.

We have never had to have a strict lockdown.

The restaurants have never been closed.

We can go out and shop.

We are, we do sensible social distancing, and we wear a mask.

And we have not had.

And, and, and to, to this day, we have had a grand total of four deaths.

You’re talking about a place with a population of seven and a half million, approximately New York city.

De Kai: and so, you know, and in total, we’ve only had a couple of hundred local transmission cases.

Almost all of our key cases came from.

De Kai: imported cases from people coming back from Europe or the U S or other places imported.

We, we got tested and, and quarantined.

De Kai: and so, De Kai: I think there are a lot of things in talking about AI ethics, right?

There are a lot of things that we can do, De Kai: with AI, De Kai: to really have a positive impact on society as a whole.

And that’s an excellent example.

Xerxes Voshmgir: That’s very interesting.

And, Xerxes Voshmgir: There there’s a website up where you can

De Kai: Absolutely.



The, the best place to start is that explainer video.

It’s only four minutes long.

De Kai: and you can find it easily at dek.ai.

That’s my name with the.in it.

De Kai: slash mask video.

So dek.ai/maskvideo,

Xerxes Voshmgir: Okay.

All right, then I’d also put it into the show notes so

De Kai: super.


In, in the video description, there are links to the rest of it.

There’s, De Kai: a website.

De Kai: with all the resources, dek.ai/masks for all, De Kai: number four, De Kai: there are, there’s a white paper that is, De Kai: highly visual and easy to read.

There’s a research paper.

De Kai: there are a number of other related videos, De Kai: list of news stories, De Kai: from around the world.

Xerxes Voshmgir: Okay.

Xerxes Voshmgir: And I’m, I’m very much interested in the topic as you also talked about, Xerxes Voshmgir: before ethics and, Xerxes Voshmgir: artificial intelligence.

Part of my studies was also philosophy.

And when I talk about artificial intelligence, mostly from the philosophical point of view, and I’m reading the example of AlphaGo Zero.

Xerxes Voshmgir: I, I got the impression, like once the artificial intelligence only needs the rules, then, Xerxes Voshmgir: it’s really important, which rules we give or programming to the artificial intelligence.

And, Xerxes Voshmgir: and, Xerxes Voshmgir: and then my personal responses, I mean, there’s the three rules of robotics, of course.

And then I go a step further.

Just me and a lot of people do to say, well, maybe we have to look at the big religions to see what the rules are that we should program into artificial intelligence.

So my question is not now, if you believe, if that’s true or not, my question goes rather than the direction of, of do you think it makes a difference if we program ethical walls into the artificial intelligence and then it’s a black box.

And we don’t know what happens because for example, when I talk to Roman Yampolskiy, the way I understood him at least was it does make a real difference because moment it reaches superintelligence,

Xerxes Voshmgir: What’s your perspective on that?

Do you think it makes a difference, which type of, Xerxes Voshmgir: ethical rules.

So to say we program into the artificial intelligence or when it’s calm, when once it becomes super intelligence, it will do its own thing.

Anyways, it doesn’t matter what we program is with beforehand.

De Kai: I don’t even think it’s possible to program it in, De Kai: you know, honestly, De Kai: The kinds of systems that you can program ethical roles into are not even true AI.

The they’re very, very simple systems.

And you know, today there’s a tendency, De Kai: you know, it’s a marketing tendency for everybody to label what they’re building as AI.

De Kai: yeah, it might be just some if then roles, De Kai: Real AI is machine.

It includes learning capability.

The machines are learning in the same way that human children learn.

And the three laws of robotics don’t really make any sense in that context, because even when you have only three rules, they’re constantly in contradiction.

To each other.


You know, Isaac Asimov wrote dozens of stories and highly entertaining stories.

And every single one of them was based on a situation where gee, these two rules create a logical conflict.

And imagine now, if you try to write an AI system that has several hundred ethical rules in them, I mean, the robot will just be paralyzed because it will constantly be in the state of contradiction.


So, how did human children manage this?

They learn right from childhood.

You’re taught values, you’re taught, De Kai: principles, but you’re also learn tradeoffs, De Kai: and you learn the difficult situations.

De Kai: and it’s highly cultural, you know, De Kai: there are a few, there are a few ethical rules that are probably universal, like, De Kai: The golden rule, you know, almost every culture has some form of the idea that you should, you should treat others the way that you would like them to treat you.

But other than that, different cultures have, you know, a lot of variation, De Kai: and the language, you know, one of the things that makes machine translation hard is it turns out that the language encodes this, De Kai: very deeply.

So some of the earliest words that children learn already encapsulate highly culture dependent values.

De Kai: and you can’t even translate three-year-old language because of that.

De Kai: it’s, it’s really tough to do.

And, De Kai: and because of that, we have, we have a situation where.

The ethical rules that an AI would have to operate with would be just as culture dependent, just as contextual, just as shades of gray as a human.


They would be learning those things.

They would not be something that you can program in, but, you know, modern AI by definition learns.

And so you can no more, you cannot program.

Ethical rules into a real AI, any more than you can hard wire them into a child, right?

There’s no way you can unscrew the head of a child and connect in solder in some logical rules.

There’s no place to connect them in that architecture.

A modern AI is the same.

So we are going to have to, as a science society, come to grips with the fact that we have to teach our AI.

In the same way that we teach the, De Kai: humans, De Kai: what are the ethics?

What are the values?

And we have to set examples because they’re learning from us.

They’re, you know, the AIS at Google, the AI’s at Facebook as a Twitter.

The AI is at Apple as a Microsoft, Amazon.

They are learning by watching what you do.

You are the training data.

You are the example.

You are the one teaching.

The ethics and the values to the AIS and we all are.

And ultimately just as with humans, there is no substitute for the fact that all of us bear collective responsibility for teaching the AIS in the same way that we do with humans.

De Kai: this is one of those things that.

Because AI has come upon society so rapidly.


We are not wrapping our heads around this.

We are still thinking about machines as dumb, passive mechanical tools as slaves.

And we don’t realize that everything changes when machines actually have opinions and thoughts and can influence our opinions and thoughts.

The metaphor of treating machines as mechanical slaves that we can control and program is a dangerously, obsolete metaphor that we need to get humanity to throw away as fast as we can.

Xerxes Voshmgir: Okay.

So, if I understand you correctly, in a way, it means that we, as a humanity need to become a role model for the artificial intelligence is as parents need to be role models or are role models for their children.

And as a parent, you cannot tell your children to not to this or not.


I may be on the short term.

You can, but you cannot implement it.

And in the same way, we, as a humanity cannot implement certain, Xerxes Voshmgir: rules and paradigms into artificial intelligences, but rather lead by example, which then if I understand this correctly, the way you say it means that we actually have to perhaps even make a quantum leap in our development as humanity.


That artificial intelligence will develop in a healthy way.

De Kai: absolutely.


And that’s what my TEDx talk, De Kai: was it, it was asking challenging the audience to ask how, how, how was your parenting?


And it turns out that when you actually dive in and look at it, we are violating pretty much every principle of good parenting that we all.

Believe in, and that’s why AI powered tech today is it is tearing apart.

Our societies, if we all abandoned responsibility for raising human children, the society will fall apart.

And that is what we’re doing now with our artificial children.

And that’s, what’s so dangerous.

Xerxes Voshmgir: Okay.

So what do you think, Xerxes Voshmgir: then the paradigms that need to be challenged in the field of artificial intelligence, but perhaps it’s even broader than artificial intelligence for us as humanity too.

Develop in a healthy way and, Xerxes Voshmgir: to develop, Xerxes Voshmgir: artificial children in a healthy way.

De Kai: Yeah, to me, that is the paradigm that we need to challenge and to shift, this is a change that is already upon us.

Like this happened while we weren’t watching it snuck up behind us.

And, De Kai: we are still living as if we were in the 20th century.

We are still, you know, pretending that things can go on.

A lot of AI, ethics is still, I think, unconsciously trapped in a mindset of thinking, how can we preserve the status quo in society?

How can we design AI so that it doesn’t.

Change the structure of society.

And I think that that is impossible even with weak AI is today.

We are seeing how impossible it is that even if we just look at the news from the last week, De Kai: all over, you know, Twitter, finally deciding that they are going to slap labels on what they consider to be.

De Kai: dangerous or false statements, even from, De Kai: a us president and the pushback on that Facebook still refusing to do that.

De Kai: we, we are already today living in a situation where we are grappling with the consequences of the fact that the AIS are already deciding based on imitating humans, learning from human behaviors.

Which ideas to spread and which ideas to sensor to filter out.


You know, I, I don’t want to like, try to be sensitive about the wording here, but an AI that is working for a social media or search engine company, its job is to decide what are the top 10 or top 20 posts or hits to display to you?

You don’t get to see the other billion 99.99% of people never look past the first 20 options that is in reality, de facto algorithmic censorship.

And there’s no way to get around that.


You have a lot of useless discussion saying, Oh, AI’s, shouldn’t be censoring.

It’s impossible because you cannot read a billion posts.


That algorithm is going to decide what you don’t see.

And so AIS are already by necessity, the P the deciders of what you do, and don’t see, and how we teach those AIS.

It’s not a question that we can continue to avoid if we want our societies to survive.

That is the major paradigm shift that needs to happen, not just among tech developers.

That is a major paradigm shift that needs to happen for every member of 21st century societies.

Xerxes Voshmgir: Okay.


And so is there anything more you would like to elaborate on the structures that you talk about, that how this structurally effecting us?

So you said the way we approach the, I want to maintain the social status quo.

So, Xerxes Voshmgir: do you have an idea which direction it ideally goals, how artificial intelligence will fundamentally change is the utopia that you see or is it just that you can say, well, the way it develops now is not healthy and, Xerxes Voshmgir: we need to rethink and, Xerxes Voshmgir: but this needs to be a creative process, Xerxes Voshmgir: where there is no.

Xerxes Voshmgir: obvious solution yet.

De Kai: So, first of all, I think the concepts of utopia and, and maybe dystopia as well are, are, those are like fairytales metaphors from certain cultural traditions.

They tend to be absolutest traditions.


They tend to, you know, the same kinds of traditions that give rise to the idea of universal truths or, De Kai: you know, that sort of absolute,

those are myths.

They don’t exist in the real world.

The real world is a process.

It’s a continual evolving flow.


And I think the sooner we realize that the better.

It’s a convenient metaphor.

We can talk about, utopia vs.

dstopia but we shouldn’t get too hung up on that because it’s not real.

What is real is how we evolve our cultures, how we evolve our civilizations, De Kai: and whether we avoid extinction, what is real is.

How much can we do away with a human suffering, which is still at unethically, large levels, given the wealth of our planet today?

I think we need to focus on

the fact that.

It’s inevitable that human intelligence and artificial intelligence is merging.

That is already happening today.

Half of my brain is already in my phone over here.

I think I can no longer remember numbers.

I used to be legendary at doing it.

I have no sense of spacial orientation anymore.

I mean, it was bad enough to begin with, but now I’m basically lost without my GPS.

De Kai: and, and.

To me.

Yeah, there’s a little bit of nostalgia.

Oh, it was nice to be able to do that, but it’s also incredible freedom because I have offloaded dreary tasks, De Kai: to the machine freeing up so much of my cognitive load capacity for creative things and De Kai: other, De Kai: aspects of, of thinking that would be great.

And this.

This synergy, this, this coupling of processing between my primitive artificial intelligence, his here in my phone and my human intelligence is the tip of the iceberg.

We are going to rapidly see in the coming decade, various forms of AR of augmented reality that will make the interface with my phone screen look clumsy.

De Kai: it will become much more instantaneous.

What I want, what I’m thinking, what I’m retrieving with, De Kai: you know, just what’s in my field of vision or in my ears.

De Kai: and that’s before we even get to, De Kai: eventually inevitably brain implants.

So all of this is going to happen.

There’s going to be older generations that resist kicking and screaming and say, Oh my God, we’re losing what it means to be human.

The younger generations aren’t going to care at all.

Let’s be honest, right?

The kids of today, De Kai: are doing things with whatever tech talk and everything else that, De Kai: you know, yeah.

De Kai: the older generations can and yell until they’re blue in the face.

And it’s not going to change what the generations do with the technology, because it is freeing and, and.

And the, De Kai: natural tendency to be nostalgic and to, to, to, you know, sort of sentimentally, want to hold on to, De Kai: what we were familiar with, De Kai: when we were kids is just not enough weight for, De Kai: the additional power of what the AI and the tech is able to give to the creativity.

Of the younger generations.

So acknowledging that and using that to rethink how we need to reform our societies is essential.

You know, like, so I’m American, De Kai: and the founding fathers in the U S 250 ish years ago.

You know, thought long and hard about how to define, define the, De Kai: the constitutional principles and, and they did a lot of what we today would call systems thinking, right.

Trying to figure out.

De Kai: all of the interplay in a complex dynamic system and how to keep balances and it, you know, sort of worked for a while.

De Kai: but it still led to an awful bloody silver war, De Kai: and the effects of which we still are grappling with even this week in the news, as everybody is saying.

De Kai: and, De Kai: the situation is being.

Made much worse spiral in a spiraling out of control by all the amplification on AI driven tech on social media, et cetera.


And so, you know, once again, it is time to rethink the way our founding fathers thought, De Kai: in the U S and in the same parallel in many other countries as well.

It is, it is actually long past time for us to rethink in a modern era with AI at the core of so much transfer of knowledge and information.

De Kai: What are the complex system dynamics?

We cannot just blindly assume that principals defined 250 years ago, or even 50 years ago still apply.

They don’t.

It’s a brave new world.

We need to do that.

Rethinking w I, you know, I don’t think we can survive doing that through another revolution that weapons of, De Kai: that we have today are far too dangerously destructive.

We’re not just talking, you know, unreliable, muskets.

We really cannot afford a revolution.

De Kai: In a hard sense to do this.

We have to raise enough public awareness to be able to understand the absolute need, to reframe the founding principles of society in an AI era.

Xerxes Voshmgir: So my last question is really.

When you imagine yourself looking back at your life from your deathbed and, Xerxes Voshmgir: everything you’ve done, looking back, what’s the impact that you want to have had on your life and on your humanity.

De Kai: I would, I would just like to have.

Made contributions to get humans to overcome their, their natural unconscious biases against having greater understanding between different peoples, between different cultures, between different neighborhoods.

De Kai: our biggest danger with all of this AI driven amplification of fear and hatred and divisiveness is.

Our own natural tendencies to divide everything into our ingroups.

And outgroups, that is a process that dehumanizes the outgroups.

If we’re going to struggle with the question of what AI, what are, what the artificial does to our society, then we need to actually really focus on.

Those biases that cause us to de-humanize too much of our society.

Xerxes Voshmgir: Okay, thank you very much.

Thank you all for your time and the interview.


De Kai: pleasure.

Xerxes Voshmgir: it was a pleasure for me.

I have it with day then.

Xerxes Voshmgir: Thank you for staying tuned for this edition of challenging paradigm acts.

If you like this episode, if they cry, feel free to share the few community.

So the guy’s message gets spread even further.

In the show notes, you’ll find links to his work and his multiple TEDx talks.

Xerxes Voshmgir: Please hit subscribe and rate my podcast.

If you like it, Xerxes Voshmgir: I’d also be add as you write me a review, you can support this podcast through Patrion.

And if you have any questions or comments, feel free to contact me next week.

We’re up with another edition of challenging paradigm X.

So until then, I wish you a great week and say cha.



We welcome you to contact us for more information
about any of our services.

Let’s Talk About A Keynote At Your Event

Xerxes offers keynotes on future technology and the future of humanity as well as tailor-made keynotes on request.

Let’s Talk About Your Future Projects

Xerxes serves as a sounding board, catalyst and innovator for your future projects.

Let’s Talk About Your Strategy

Xerxes and his team offer strategy consulting, strategic PR and organization development services.