Introduction

The purpose of this talk is to inspire.

Just a bit about me. My name is Dhruv.

I've been working in AI for about 10 years now. Worked in various different fields. I've been an entrepreneur within computer vision. Right now I run the AI team at Uber in the product team. So I've done various different things in AI.

The Future of AI

And one of the things that I've been thinking about for the last particularly few years is where are we headed? with AI. Where are we going as humans? And that's kind of where I wanted to explore this talk.

So since, arguably, the dawn of agriculture, humans have been on a relentless quest for progress. So we've grown more crops. We've built taller buildings. We've been focusing on doing more, achieving more, and being better.

Work and AI

And so one of the things I've been thinking about is that Where we're headed with AI, just to zoom out a lot, is that we're actually approaching a world where work might just disappear completely.

And so not only that, what AI is starting to do is it's starting to go into areas which used to be parts of leisure. So it's able to do things which we used to spend our leisure time on. So it can create pieces of art. It can write beautiful pieces of music. It can write scripts for films.

And so as humans, we really have to reevaluate what is the point of doing anything? Is the goal of doing something just to do it better, more productively? Or is the goal of a task just to enjoy it? Does anything have value for its own sake or do you always have to be best at everything? Is that the sort of progress of where humanity is headed?

Cool.

Risks of AI

So let's dive into the risks we currently face in this AI-driven world.

So any of you who know about Karl Marx, I mean, I studied economics. We learned about this thing called alienation, right?

And he talked about this idea that basically we would be chasing, we'd be workers in a certain factory or a a sort of institution where we feel alienated from the goal of that institution. In our world that we live in, that's obviously around growth, productivity, you're measured in your job by sort of your output.

And so this idea of this elusive concept of productivity, he envisaged a world where we would basically be alienated. We wouldn't see any purpose in what we do.

Alienation and Inadequacy

Let's think about a graphic designer who's used to pouring her heart into her project and now all she's doing is tweaking AI-generated designs.

The other risk that we have is this feeling of inadequacy. So imagine you've been spending 20 years mastering a certain skill that can be done in seconds with an AI. So take a copywriter who has 20 years of experience writing copy and watches an AI churn out that copy within seconds. Imagine the feeling of inadequacy.

So it's not just our skills becoming obsolete, it's our sense of self-worth. How do we measure our value in a world where machines can outperform us in everything?

Loss of Imagination

So let's kind of also think about the loss of imagination. If an AI can handle even creative tasks, could we actually lose all of our ability to even imagine and think outside the box and do anything?

Imagine, let's take another example, a novelist who starts using AI to generate plot ideas and character arcs in a story. And over time, all she's relying on is the AI's suggestion to create those stories, and she starts to doubt her own creative instincts.

So is she actually then just a curator, just a mere curator of AI-generated content, or is she actually a creator in her own right? Finally, losing any purpose in our actions. How can we achieve meaning?

Purpose and Meaning

Imagine team meetings now, where any small talk at the beginning of the meeting is replaced by an AI-optimized agenda, and you're not spending your weekends on your hobbies, but you're spending your weekends on an AI-assisted self-improvement course. You know, the funny thing about all of this is that the development of AI, our world, needs us as humans to progress. So if we stop creating and we stop innovating, how will the AI itself learn to evolve?

Could we be in a situation where both humans and AI end up in a plateau? Cool.

Human Essence in the Age of AI

So what's the essence of being human in this age of AI? How do we differ from machines? Let's try and explore the core arguments of what makes us human. So let's go through all of them.

So first of all, the argument is that we have empathy, right? We've all had that feeling of consoling a friend who's just lost a loved one. That ability to feel someone else's pain and offer comfort when it doesn't actually benefit us directly is actually quintessentially human. Do we think that an AI can actually experience, for example, loss?

Secondly, humans have emotional intelligence, right? So let's take, for example, a diplomat who's navigating a tense international negotiation. So for that diplomat, they're not just processing information, but they have to read subtle human cues, they have to anticipate reactions, and they have to choose their words really carefully.

So let's take a third argument about humanity. We possess boundless creativity, right? So we stem from a place as humans that comes from dreams, subconscious, our subconscious, you know, experiences that an AI probably doesn't have any ability to replicate. Think about Salvador Dali's paintings, or think about us as children seeing cardboard boxes and turning them into spaceships, for example. Could an AI do that level of imagination? That ability to create something from nothing and see potentials where others see limitations is some of the things that make us human.

The fourth argument about humanity is that we're self-aware. So when you look in the mirror and you recognize yourself, when you reflect on your actions and you feel pride and regret, you're exercising a level of consciousness beyond just mere data processing. So do we think that an AI can actually know itself in that same way?

So fifth argument, we sense, we feel, that feeling of biting into an apple, the feeling of the sun on your face, the feeling of swimming in a lake, these sensory experiences combined with our memories are something that probably an AI, which is a disembodied system, couldn't actually replicate or comprehend.

So finally, we have intentions, we have goals, and we have dreams. So think about the athlete who's been training for years to compete in the Olympics. So that driven by his or her own personal dream, does that intrinsic motivation that sort of drives our human progress is something that an AI probably doesn't have.

So I want us to think about those aspects and think about as AI continues to advance, first of all, do we think it can replicate those traits or those traits that we should be focusing on thinking about more and enhancing in ourselves?

Ensuring AI Enhances Humanity

So where should humanity focus its attention? How can we ensure AI enhances us rather than hinders our human connections?

Imagine a world where AI can handle all of our schedules and tasks, all the sort of boring automation is done of all of our lives, and it freezes up to have deeper conversations with our loved ones. A family dinner where smartphones are replaced by eye contact and laughter.

Secondly, how can AI become a tool that amplifies our creative potential and focuses on that rather than replaces it? Imagine if you had masterpieces that were created between AI and humans where the AI was bringing the sort of computational precision of the painting and the human was bringing the emotional side of that idea.

So thirdly, how can we harness AI to gain deeper insights about ourselves and our own personal growth and self-discovery? So imagine you had an AI-powered journal that wasn't just recording your thoughts and transcribing them, but actually helping you uncover patterns in your own behavior and your own emotions.

How do we redirect our focus using AI as a tool to tackle humanity's biggest challenges? Let's have AI systems working alongside humans, experts, to solve climate change and cure diseases and be a thought partner.

We have to overcome this sort of endless desire for more that I talked about at the beginning. Let's visualize a society where success isn't about productivity and output, but driven by your own personal fulfillment and contribution to the community.

So how can we use AI to redefine our worth beyond our own output? Yeah, so some thoughts there. Cool.

Redefining Work

Now that we know that work might actually disappear completely, let's actually redefine what work is. As we talked about at the beginning, the idea of work was centered on converting your time into generating output, initially from the land. Then it evolved into converting time into using tools to make more products and services for more people to buy.

In the landscape that we see right now where machines could potentially automate most product and service output from end to end, we may not need any human involvement. So what becomes the purpose of work? Why do you go into an office?

So I think my view on this is that work has to shift dramatically from just driving productivity to actually helping us, the AI should actually help us explore meaningful, engaging uses of our time. So beyond just producing things, logos, designs, documents, code, plans, what do we actually produce?

Imagine a world where you're moving from a 10-hour workday to a three-hour workday. What do you do with those extra seven hours? Do you dedicate the rest of your time to hobbies and personal growth?

Are we spending enough time in society actually thinking about what we spend our time on when we're free? I think that's a question that we should be thinking about.

Solarpunk and AI Punk

So how many of you know about the Solarpunk movement? Raise of hands.

Cool, okay, so do you guys know about Cyberpunk? Okay, so there's two sort of artistic movements, right?

One of them was cyberpunk, where this is where you sort of had dystopian futures, you've got society collapsing, dominating corporations, that was kind of one vision. Then you had steampunk, which was an aesthetic around kind of the industrial revolution, a love, a nostalgia for sort of people using machines.

And Solarpunk was a movement that sort of was on the fringes of the internet around 2000, but sort of there was a manifesto that was written around 2020 where sort of humanity could coexist with technology working with the climate. So you can see visions like this where you have sort of You know, there's no climate crisis. We're coexisting with nature. There's machines kind of running around, you know, watering the plants and flying around with drones and monitoring our tomatoes and so on.

So I think we need to think about a new type of movement, which we can just kind of forget the naming for now, but let's call it the AI punk movement. Think about this kind of vision where you sort of have us living with nature, but where we're living in harmony with AI, right? What could that look like?

So imagine where we as humans are training AIs to enhance but not replace our own human capabilities. So we're actually getting better with the AI. Imagine humans and AI forming complementary teams on projects.

What does that look like? What happens if AI was actually to suddenly not just sort of help us do certain tasks, but actually completely enhance our subconscious, our dreams, and our creativity and our imagination?

Collaborative Future

So imagine you were to walk into your home office and you had an AI assistant projected as a hologram and you just sort of came up with some ideas and thoughts for the day and it was just able to sort of come up with how to do them and execute in that day. And it sort of offers you suggestions, filling the room with sort of 3D projections of all your ideas.

So I think to actually build something like this, we have to develop uniquely, we have to develop these unique human skills that we talked about before that complement the AI capabilities. And we should be using and harnessing AI to amplify our creativity and sort of all of those problem solving capabilities that we have and form a partnership with the AI.

So I think one thing that, you know, where can AI sort of change its course from now, right? It's now very much focused on sort of automating tasks, helping us in the workplace. We saw some of those talks before, right?

But how can we help build AI that helps us find moments of joy? So imagine an AI that sort of starts to remind you to call a friend that you haven't and it knows that you can measure your own well-being and it knows that that's the right thing for you and it gives you those sort of tasks which are less about productivity and output.

Imagine sort of an AI journaling companion that sort of probes into how you're feeling and helps you sort of think about, ask these questions that probe your own self-discovery. Imagine your smartwatch didn't just track your steps, but it helped you understand your emotional patterns through the day.

You know, all the social media platforms that we built today are really about sort of effectively our own egos, right? You know, they help us connect with each other, but how could they completely change, right? How can social media platforms, for example, encourage empathetic interactions?

So yeah, I think this kind of gives you a sense of kind of, I think one interesting point here as well is like how, Some of the challenges that we face in humanity are also because we've built a world where the systems that we have are very much around self-interest, right?

We have a goal function that we're trying to achieve. And a lot of the, if you think about any AI, it has a goal function, it has an optimization function. And you're trying to compute the loss of that and trying to achieve this outcome. And they're getting better and better at doing that.

One of the benefits of AI compared to human beings who have these selfish goals is that potentially AI can actually step back and coordinate amongst these different groups that are trying to achieve a certain outcome and actually think in a very objective way about these really hard problems. So for example, think about designing a city. You know, how can you design a city that maximizes community well-being? You know, are we the best placed as humans to do that, right, given the self-interest that we have?

Cool. Let's kind of move on. So...

Future Steps

What should we do going forward? How can governments and institutions change how they're thinking about this?

I think the first thing that we need to change, and I've thought about this for a while, is our measure of output in the world, and you think about GDP, is literally the sum of our expenditures, or it's the sum of the wages or rent or income of people, or it's the net value of production. So in a country like Bhutan, for example, people go around with surveys and they ask people how happy they are. It's very interesting what questions they ask.

How can we all move towards systems where we actually truly judge where we're headed beyond just the output of our own production? How do we also measure, talking about how important it is for AI to actually have human input and new creativity and ideas being created, how do we actually measure that we're actually putting new inherent knowledge into the system? You know how much actual ideas are we putting into the system that the sort of new intellectual capital? Like are we measuring that in a proper way?

So How do we also change education? You know, how do we change? You know you know, classrooms where children aren't just memorizing facts and knowledge, which we've kind of, we're going to achieve, right? But how do we actually teach them to be more empathetic, creative, and adaptable? Like, is that where the education system is actually helping us?

And then sort of think about, you know, going back to that sort of AI punk vision, right? Imagine public spaces where, you know, you have AI generated art and human creations kind of coexisting, you know, sparking conversations about that interplay. Cool.

Conclusion

So I'm going to wrap up the talk. So three key points.

I think we need to actively shape AI's role to preserve our own human essence. And that means developing AI systems that enhance rather than replace, you know, our uniquely human qualities that we talked about.

Secondly, we need to prioritize human well-being more than ever in this age. It's not just about what AI can do, but how it can improve our own lives and improve society as a whole.

And third, and most importantly, we have to celebrate and nurture what makes us uniquely human, which is our creativity, our empathy, and our ability to connect on a deep emotional level. And that's what I'd urge us to start focusing on for the next 10 years.

Thank you.