The AI-powered tools supercharging your imagination | Bilawal Sidhu

[00:00:00] TED Audio Collective.

[00:00:08] There's a fine line between AI replacing us and AI serving as a tool to further elevate

[00:00:16] our work.

[00:00:18] So instead of asking AI to do our work for us, what if we took a step back and thought

[00:00:23] much more consciously about AI's capabilities and opportunities?

[00:00:28] Content creators and technologists like Bilal Bou Siddiou don't see AI as a threat to our

[00:00:32] jobs or new tech limited to eliminating the mundane.

[00:00:36] In his view, it's more of a workhorse for unlocking our imaginations and setting our

[00:00:41] creativity on fire.

[00:00:43] In sentiments he shared in a tweet, Bilal said,

[00:00:46] Don't worry y'all.

[00:00:48] AI will do your laundry and your dishes so that you can do your art and writing, not

[00:00:52] just the other way around.

[00:00:56] I'm Sherelle Dorsey and this is TED Tech.

[00:01:02] On the show today, Bilal Bou Siddiou takes to the TED stage to show us exactly how reimagining

[00:01:07] the world through AI might lead us into a much more exciting future.

[00:01:12] He reflects on the future of AI and his latest endeavors as the host of our new podcast,

[00:01:17] The TED AI Show.

[00:01:18] Take a listen and then stick around for my interview with Bilal Bou.

[00:01:29] This show is brought to you by Schwab.

[00:01:32] You're here because you like to keep a pulse on trends in technology.

[00:01:36] Well, now you can invest in what's trending in artificial intelligence, big data, robotic

[00:01:42] revolution and more with Schwab investing themes.

[00:01:47] It's an easy way to invest in ideas you believe in.

[00:01:50] Schwab's research process uncovers emerging trends.

[00:01:54] Then their technology curates relevant stocks into themes.

[00:01:58] Choose from over 40 themes.

[00:02:00] Try all the stocks in a theme as is or customize to better fit your investing goals.

[00:02:07] All in a few clicks.

[00:02:09] Schwab investing themes is not intended to be investment advice or a recommendation of

[00:02:14] any stock or investment strategy.

[00:02:17] Learn more at schwab.com slash thematic investing.

[00:02:23] This episode is brought to you by Progressive.

[00:02:25] Are you driving your car or doing laundry right now?

[00:02:28] Insurance go best when they're bundled with another activity.

[00:02:31] Like Progressive home and auto policies, they're best when they're bundled too.

[00:02:35] Having these two policies together makes insurance easier and could help you save.

[00:02:40] Customers who save by switching their home and car insurance to Progressive save nearly

[00:02:43] $800 on average.

[00:02:46] Quote a home and car bundle today at progressive.com.

[00:02:50] Progressive Casualty Insurance Company and Affiliates.

[00:02:52] National average 12-month savings of $793 by new customers surveyed who saved with Progressive

[00:02:58] between June 2021 and May 2022.

[00:03:01] Potential savings will vary.

[00:03:04] Hackers and cyber criminals have always held this kind of special fascination.

[00:03:08] Obviously, I can't tell you too much about what I do.

[00:03:12] It's a game.

[00:03:13] Who's the best hacker?

[00:03:14] And I was like, well, this is child's play.

[00:03:17] I'm Dena Tumbler-Rustin.

[00:03:18] And on the Click Here podcast, you'll meet them and the people trying to stop them.

[00:03:22] We're not afraid of the attack.

[00:03:24] We're afraid of the creativity and the intelligence of the human being behind it.

[00:03:29] Click here.

[00:03:30] Stories about the people making and breaking our digital world.

[00:03:33] AI machines.

[00:03:34] Satellites.

[00:03:35] Engine ignition.

[00:03:36] Click here.

[00:03:37] And lift off.

[00:03:39] Click here every Tuesday and Friday, wherever you get your podcasts.

[00:03:43] All right, let's talk about blending reality and imagination.

[00:03:48] But first, let's take a step back in time to 2001.

[00:03:53] As an 11-year-old in India, I became obsessed with computer graphics and visual effects.

[00:03:58] Of course, at that age, it meant making cheesy videos, but therein started a foundational

[00:04:03] theme in my life, the quest to blend reality and imagination.

[00:04:08] And that quest has stayed with me and permeated across my decade-long career in tech, working

[00:04:13] as a product manager at companies like Google and as a content creator on platforms like

[00:04:19] YouTube and TikTok.

[00:04:21] So today, let's deconstruct this quest to blend reality and imagination and explore how

[00:04:27] it's getting supercharged, buzzword alert, by artificial intelligence.

[00:04:32] Let's start with the reality bit.

[00:04:35] You probably heard about photogrammetry.

[00:04:37] It's the art and science of measuring stuff in the real world using photos and other sensors.

[00:04:44] What required massive data centers and teams of experts in the 2000s became increasingly

[00:04:50] democratized by the 2010s.

[00:04:53] Then of course, machine learning came along and took things to a whole new level with

[00:04:57] techniques like neural radiance fields or NERFs.

[00:05:01] But unlike older techniques for reality capture, NERFs do a really good job of encapsulating

[00:05:08] the sheer complexity and nuance of reality, the vibe, if you will.

[00:05:14] Twelve months later, you can do all of this stuff using the iPhone in your pocket using

[00:05:19] apps like Luma.

[00:05:20] It's like 3D screenshots for the real world.

[00:05:23] Capture anything once and reframe it infinitely in post-production.

[00:05:28] So you can start building that collection of spaces, places, and objects that you truly

[00:05:32] care about and conjure them up in your future creations.

[00:05:37] So that's the reality bit.

[00:05:39] As NERFs were popping off last year, the AI summer was also in full effect with MidJourney,

[00:05:46] Dolly 2, Stable Diffusion all hitting the market around the same time.

[00:05:51] But what I fell in love with was in-painting.

[00:05:54] This technique allows you to take existing imagery and augment it with whatever you like

[00:05:59] and the results are photorealistically fantastic.

[00:06:03] It blew my mind because stuff that would have taken me like three hours in classical workflows

[00:06:08] I could pull off in just three minutes.

[00:06:12] But I wanted more.

[00:06:13] Texture Control Net, a game-changing technique by Stanford researchers that allows you to

[00:06:18] use various input conditions to guide and control the AI image generation process.

[00:06:25] So in my case, I could take the depth information and the texture detail from my 3D scans and

[00:06:31] use it to literally reskin reality.

[00:06:34] Now this isn't just cool video, there's a lot of useful use cases too.

[00:06:38] For example, I'm taking a 3D scan of my parents' drawing room as my mother likes to call it

[00:06:44] and reskinning it to different styles of Indian decor and doing so while respecting the spatial

[00:06:51] context and the layout of the interior space.

[00:06:55] You could take that 2016 scan of a Buddha statue and reskin it to be gloriously golden while

[00:07:01] pulling off these impossible camera moves you just couldn't do any other way.

[00:07:06] Or you could take that vacation footage from your trip to Tokyo and bring cherry blossoms

[00:07:11] to life in a whole new way.

[00:07:13] And let me tell you, cherry blossoms look really good during the day but they look even

[00:07:17] better at night.

[00:07:20] It's almost like this dreamlike quality where you can use AI to accentuate the best aspects

[00:07:25] of the real world.

[00:07:27] Natural landscapes look just as beautiful, but of course you could go over the hills and

[00:07:31] far away to the French Alps from another dimension.

[00:07:36] And it's not just static scenes, you can do this stuff with video too.

[00:07:40] I can't wait till this technology is running at 30 frames per second because it's going

[00:07:45] to transform augmented reality and 3D rendering.

[00:07:48] I mean how soon until we're channel surfing realities layered on top of the real world?

[00:07:55] Of course, just like RealityCapture got democratized, all these tools from last year are getting

[00:08:00] even easier.

[00:08:01] So instead of me spending hours weaving together a bunch of different tools, tools like Runway

[00:08:07] and Kyber let you do exactly the same stuff with just a couple clicks.

[00:08:11] Want to go from day to night?

[00:08:13] No problemo.

[00:08:14] Want to get that retro 90s aesthetic from full house?

[00:08:18] You can do that too.

[00:08:20] But it goes beyond RealityCapture.

[00:08:23] Companies like Wonder Dynamics are turning video into this immaculate form of performance

[00:08:28] capture so you can embody fantastical creatures using the phone into your pocket.

[00:08:34] This is stuff that James Cameron only dreamt about in the 2000s and now you can do it with

[00:08:39] your iPhone?

[00:08:40] That's absolutely wild to me.

[00:08:42] So when I look back at the past two decades and this ill-tailored tapestry of tools that

[00:08:48] I've had to learn, I feel a sense of optimism for what lays ahead for the next generation

[00:08:53] of creators.

[00:08:55] The 11-year-olds of today don't have to worry about all of that crap.

[00:09:00] All they need to do is have a creative vision and a knack for working in concert with these

[00:09:04] AI models, these AI models that are truly a distillation of human knowledge and creativity.

[00:09:11] And that's a future I'm excited about, a future where you can blend reality and imagination

[00:09:16] with your trusty AI copilot.

[00:09:19] Thank you very much.

[00:09:26] That was creative technologist Bilal Wasedu speaking at TED 2023.

[00:09:34] Bilal is also the host of our new podcast, The TED AI Show, and we are delighted to have

[00:09:39] him on the show today to talk more about this exciting technology.

[00:09:43] Bilal, it's so great to speak to you today.

[00:09:46] I love your show.

[00:09:47] I've loved your TED Talk.

[00:09:49] This is really exciting to have conversations about technologies transforming the creator

[00:09:54] economy.

[00:09:55] You've been in this game for a really, really long time.

[00:09:57] How are you doing today?

[00:09:58] I'm doing great, Cherelle.

[00:09:59] Thank you so much for having me.

[00:10:01] So one of my first questions for you is, how do you see some of these new tools that you

[00:10:07] talk about all the time?

[00:10:08] I follow you online.

[00:10:09] It's amazing.

[00:10:11] How do you see these tools and technologies transforming the creator economy?

[00:10:16] I think if I had to put it in a pithy sentence, I would say indies will be able to rival the

[00:10:21] output of studios, and then studios are going to set whole new standards altogether.

[00:10:25] So movies will be made that wouldn't have been otherwise.

[00:10:28] Games will be made that wouldn't have been otherwise.

[00:10:30] Podcasts, TV shows, music, the list goes on.

[00:10:33] So we'll have this explosion of content that just would not have made it through the traditional

[00:10:37] green lighting process.

[00:10:40] But this transformation will also impact how the work gets done.

[00:10:43] So I think like smaller teams and even solo creators will be able to punch way above their

[00:10:47] weight class.

[00:10:49] A common story that you might relate with as well is like creatives, as they grow their

[00:10:53] audience, they start building a business around it.

[00:10:55] They start adding people, processes.

[00:10:58] And the next thing you know is like you're managing all that overhead.

[00:11:01] You're looking at fricking status reports, sitting in a bunch of meetings.

[00:11:05] And so I think creatives who really enjoy making and creating will be able to do a lot

[00:11:09] more with less and retain that spark as they scale.

[00:11:14] Now on the other end, I think big teams will be able to break boundaries.

[00:11:18] So every time we get these new creation superpowers, we don't just pull a Tim Ferriss four hour

[00:11:23] work week and sit pina coladas on the beach.

[00:11:25] No, like instead we're going to put more value on the screen.

[00:11:29] So can you imagine what Disney or Marvel will be able to do with their treasure trove of

[00:11:34] IP and really visual umami?

[00:11:37] They're going to raise the ceiling and set whole new standards altogether for what cinematic

[00:11:41] experiences can do.

[00:11:44] I love that.

[00:11:45] I love that.

[00:11:46] And, you know, in your in your previous talks, you've also talked about how some of these

[00:11:50] technologies in the past have really been clunky and disjointed or limited.

[00:11:55] And now we're sort of in this space where AI is really allowing us to expand that creative

[00:12:00] imagination of new worlds.

[00:12:02] I mean, spoke about, you know, the opportunities that, you know, folks like Disney and Pixar

[00:12:07] might be able to have.

[00:12:08] And now how much more frictionless some of these AI tools have become as they've advanced.

[00:12:13] So I think for the big folks or the big guys, as well as those who are individual creators,

[00:12:18] you know, what skills will users need now to access and leverage these tools?

[00:12:23] Are they differing greatly from previous tools?

[00:12:26] It's a great question.

[00:12:27] I think the skill that's going to be paramount in this new age is basically good taste and

[00:12:33] the ability to articulate your vision.

[00:12:35] I think that's going to become a lot more important.

[00:12:37] In other words, the craft of creation is absolutely changing.

[00:12:41] It's almost worth answering your second question first about how these tools differ from previous

[00:12:46] generations.

[00:12:47] So taking a step back thus far, we've been creating at this lower level of abstraction.

[00:12:52] We're painting the pixels, we're wrangling the geometry, we're cutting the frames of

[00:12:56] video and now with generative AI, we're describing what we want to create at this higher level

[00:13:01] of abstraction.

[00:13:03] So it's almost like going from playing every instrument in the orchestra to becoming the

[00:13:07] conductor of the symphony.

[00:13:09] So in this world where we're creating at these progressively higher levels of abstraction,

[00:13:14] I think the ability to articulate what's in your head in different mediums like text,

[00:13:19] voice, visual, spatial, etc.

[00:13:21] I think that's going to be central.

[00:13:24] In many ways, this is not too dissimilar from how a creative director would prompt a group

[00:13:29] of humans to create something and then iterate with them to get to that final product.

[00:13:34] I think the other big skill, which is probably cliche and like oversaid, but it's true, is

[00:13:39] adaptability.

[00:13:41] I think this era of having one specialized skill set and writing it for your whole career

[00:13:46] is gone.

[00:13:48] Everything will keep changing.

[00:13:49] So I think it's going to be very important to cultivate this sort of relentless curiosity

[00:13:54] to keep learning in order to thrive in this new world.

[00:13:57] So I believe we're going from a world of specializations to a world where generalists will thrive.

[00:14:02] As long as creatives don't get too attached to the craft that gets us to that end, I think

[00:14:07] we'll have an amazing explosion in creativity and that will only go up.

[00:14:11] I really love that.

[00:14:12] And that really begets my next question here because we've hyper-focused, I think, on the

[00:14:17] creator economy.

[00:14:19] But that kind of immediately makes me think of entertainment and art or what have you.

[00:14:25] But what other industries do you see being transformed by better AI tools?

[00:14:31] I think every industry that involves knowledge work or really doing anything on a frickin'

[00:14:35] computer will be transformed.

[00:14:38] So I think it's the same with any industry relying on audiovisual creation, which is we're

[00:14:42] talking about entertainment.

[00:14:44] Now whether you're an architect that's using Autodesk tools or creative that's using Adobe

[00:14:48] tools, a marketer that's using Canva or heck, you're in finance and accounting and using

[00:14:53] Microsoft Excel, it really doesn't matter because AI capabilities are being infused

[00:14:59] to the tools that you use every single day.

[00:15:01] At the same time, alternatives are popping up that reimagine that entire workflow.

[00:15:07] So I think the impact is going to be pervasive.

[00:15:09] I spent four years leading 3D efforts on the Google Maps team and we were basically building

[00:15:14] up this ground-up 3D model of the world, sort of remapping the world, if you will, using

[00:15:19] all these very expensive fancy sensors like satellites, the cameras on airplanes, Street

[00:15:23] View cars. And now those technologies are far more democratized than ever before and

[00:15:29] really even rivaling the precision and accuracy of those high-end modes of creating maps.

[00:15:34] The speed is insane, too.

[00:15:36] I mean, in my TED talk, I was featuring 3D scanning with your smartphone.

[00:15:40] And even a year ago, some of those techniques relied on data centers like some Nvidia GPUs

[00:15:45] cranking on my scan and giving me back something.

[00:15:48] And now just 12 months later, it can literally happen on the iPhone in your pocket with apps

[00:15:53] like Scanniverse. So literally before I'm done uploading my scan to process in the cloud,

[00:15:58] it's done processing on my phone.

[00:16:01] I think beyond that creativity and sort of art and play, why do you feel that these advanced

[00:16:07] technologies that are deepening our understanding and our ability to really reimagine that

[00:16:12] real world are necessary?

[00:16:16] Yeah, and it's an interesting question because when people hear a phrase like that, they

[00:16:20] think like a ready player one, right?

[00:16:22] So I mean, I wouldn't even say it's about reimagining.

[00:16:25] It's almost like a reconnecting more deeply and intuitively with the world around us,

[00:16:30] the real world around us, the people within it, the places, the spaces and communities

[00:16:35] we care about.

[00:16:37] If we look at what we do today as humans right now, we spend a bunch of our time in the

[00:16:41] digital world and the physical world.

[00:16:43] Our identities and activities are kind of sharded.

[00:16:47] I think we finally have this technology to start connecting those bits and atoms and

[00:16:51] build a bridge between the physical and digital world.

[00:16:54] And so why do I think this is important?

[00:16:57] I think this is important because tech thus far has been great, but it's also been a

[00:17:02] tremendous distraction.

[00:17:04] Like these slabs of glass in our pocket mediate our interactions with the real world.

[00:17:09] But with the rise of AI and spatial computing or AR VR, I think we can start connecting the

[00:17:15] physical and digital worlds in a way in which computing can be there when you need it, but

[00:17:20] it can also get out of the way when you don't.

[00:17:23] Something about that that's deeply exciting to me is that we've suddenly got technology

[00:17:27] that can adapt to us versus us adapting to it.

[00:17:31] I love your point about technology and distraction versus, you know, us, us allowing this technology

[00:17:38] to be adaptive to our needs.

[00:17:41] And I think a lot of times we kind of get lost in the novelty of the actual tool itself

[00:17:46] versus now how do we leverage this to kind of enhance ourselves and really apply this

[00:17:53] to the work that we're doing, be it creative or knowledge work, which I think kind of goes

[00:17:57] back to some of your earlier points.

[00:17:59] So Bilal, before we go, I'd like to get your thoughts on, you know, becoming a podcast

[00:18:04] host for a show all about AI.

[00:18:07] And as you've been making the TED AI show, what makes this show and your perspective

[00:18:11] on AI different from whatever else is out there now?

[00:18:15] Yeah, you know, it's like I like to joke is, you know, what the world needs is another

[00:18:19] AI podcast.

[00:18:22] And so, I mean, what makes this show different and what has me excited?

[00:18:25] I would say, like, I'm coming at this from a couple of different lenses as a product

[00:18:29] builder, as a creator myself, as an early stage investor, and of course, as like just

[00:18:34] an exceedingly curious person who wants to make sure tech to our previous point has this

[00:18:38] symbiotic relationship with humans and society at large.

[00:18:42] And so there's a lot of dual use technologies that are being developed right now, and there's

[00:18:46] a lot of hard questions to be answered.

[00:18:49] And so I view this show as an opportunity to engage with the people that are building

[00:18:53] the future and are impacted by the future and asking those hard questions so we can

[00:18:58] think about sort of the dual futures of AI, the desirable futures, the undesirable futures,

[00:19:03] and hopefully shift us towards a future that is largely or disproportionately positive for

[00:19:08] humanity.

[00:19:09] I don't think I'm not naive enough to think that it'll be perfect and will create this

[00:19:13] utopia.

[00:19:14] But I also think this whole space is moving so fast that it can be overwhelming.

[00:19:18] So think of this show as a weekly guide for you to live and thrive in a world where AI

[00:19:23] and really technology is changing everything.

[00:19:26] In other words, how do we navigate AI before it navigates us?

[00:19:29] I love that.

[00:19:30] I love that.

[00:19:31] I know for one, explaining what's happening in AI to my friends and family is a daunting

[00:19:39] task.

[00:19:40] So I will definitely be pushing folks and our listeners as well to subscribe to the

[00:19:46] TED AI show.

[00:19:48] And I'm so grateful that you joined us here on TED Tech.

[00:19:52] Thank you so much for having me.

[00:20:01] TED Tech is part of the TED Audio Collective.

[00:20:03] This episode was produced by Nina Lawrence, edited by Alejandro Salazar, and fact-checked

[00:20:09] by Julia Dickerson.

[00:20:11] Special thanks to Maria Ladius, Ferde Grange, Corey Hajim, Daniela Valareso, and Michelle

[00:20:18] Quint.

[00:20:19] I'm Sherelle Dorsey.

[00:20:20] Thanks for listening and talk to you again next week.