Speaker A
That's really good to be here. It's nice to see all of you. I prepared my remarks.
Andrew Ng and David S. Goyer discuss AI's creativity, intelligence, risks, and societal impact, balancing hype and realistic views.
Ask about this video. Answers come from its transcript only — with the timestamp, so you can check them.
Generated from the transcript and can be wrong — check the timestamp.
Key Takeaways
- AI hallucinations occur but are less frequent than commonly perceived.
- Fear-driven AI narratives can hinder technological progress and adoption.
- AI offers significant benefits in creative and professional workflows.
- Balanced regulation is crucial to foster innovation while managing risks.
- AI-human collaboration will shape the future of creativity and work.
What the video covers
- Andrew Ng and David S. Goyer explore the evolving narratives around AI, addressing both doomsaying and utopian views.
- They discuss AI hallucinations, their frequency, and how AI models have improved over recent years.
- The speakers critique sensational media coverage that exaggerates AI risks like blackmail or human extinction.
- They emphasize the damaging effects of fear-based narratives on AI adoption and innovation in America.
- David S. Goyer shares personal experiences using AI in creative workflows like concept art and writing.
- The conversation touches on AI's potential to displace jobs and the balance between benefits and risks.
- They highlight the importance of sensible regulation and the challenges posed by state-level AI policies.
- Open-source AI models and global efforts to control AI technology access are discussed.
- The speakers reflect on the future of AI-human collaboration, especially in creative industries like Hollywood.
- They conclude with a nuanced perspective on AI's role in augmenting human creativity rather than replacing it.
Chapters
- 00:00Introduction and AI Avatar Experience
- 03:04Framing AI Narratives: Dangers and Hype
- 06:04Impact of Fear Narratives on AI Adoption
- 09:26AI in Creative Workflows and Hollywood
- 12:37Regulation and State-Level AI Policies
- 19:06Open-Source AI and Global Technology Control
- 29:09Future of AI-Human Collaboration
- 35:17Creative Use Cases and Closing Thoughts
Full Transcript — Download SRT & Markdown
Speaker A
The only thing I had AI do, well, I had AI do two things. I had AI organize the remarks into a flow and then also, on your website, you have a trained avatar of yourself to answer
Speaker A
questions. So I asked your avatar all these questions ahead of time, and I'm just curious to see to what extent your avatar is going to answer the way that you're going to answer. So, but it's not a gotcha moment. Anyway, I just thought
Speaker A
that was interesting. I was like, your avatar was there and it said, "Talk to me." So, I said, "Okay, well, this is a great opportunity." It's interesting. You know, we worked on that avatar for a long time, like a year
Speaker A
and a half now. And it was really bad initially. It kept on saying that I was I would never say that.
Speaker A
And then over time, it got better and better. It still says some weird stuff that I'm not happy with.
Speaker A
But one interesting thing I learned is when people talk to my avatar, my avatar makes mistakes and hallucinates.
Speaker A
One of my internal team members said, "I talked to Avatar and it told me about this great mountain climbing thing you did." I said, "I've never climbed those mountains, but people still seem satisfied talking to my avatar." I can't remember. I think it was
Speaker A
ChatGPT when I was—it's the equivalent of Googling yourself, but I was asking ChatGPT just to give me a quick bio of myself, and it said that I was also a survivor of pancreatic cancer, and I've never had
Speaker A
cancer, much less pancreatic cancer, and so I was like, where the hell did that come from? Yeah, they do, they do still hallucinate. What, since we're talking about this, on average, do you know what percentage of the time the current models are
Speaker A
hallucinating? Yeah, I don't know. They definitely do hallucinate, but probably less than most people think. Two, three years ago, AI models were much less advanced than they are today. And there were some widely publicized examples of
Speaker A
AI models hallucinating, like, you know, the lawyer that got into trouble for made-up court filing details and so on.
Speaker A
The technology has gotten much better, and it still does make mistakes, but probably less than most people think.
Speaker A
Great. Okay. So, I have a couple of prepared questions. We'll get into a conversation. You may have some questions for me, and then hopefully we can take some questions from the audience. I just wanted to start with a
Speaker A
really broad framing question that we had a Zoom a week or so ago. So, we had a kind of pregame, and I now know which side of the fulcrum you stand on, but there are
Speaker A
AI doomsayers and sort of AI utopians. And so, where on that spectrum would you say you fall?
Speaker A
The boring answer, I'm actually neither of the above. I think I'm much more in the middle. The way that AI narratives have evolved over the last few years has been kind of weird, and there's been a lot more doomsaying or a
Speaker A
lot more fear narratives about AI being very destructive than most people would have expected. The funny thing that happened was a handful of companies decided that stoking fear about AI was wonderful PR because it turns out that if you go [clears throat] around and say
Speaker A
my technology is so powerful I could cause the extinction of the human race, that turns out to be one way to convince people your technology is really powerful and therefore must be worth a lot. After that narrative fell apart,
Speaker A
the next one became, now my technology could take all of your jobs, so how much is my technology worth? Got to be a lot, right? And that narrative is slowly falling apart. Oh, and then a more recent one, my
Speaker A
technology is so smart it could blackmail people. We've seen lots of stories about that.
Speaker A
Yeah. And it's been interesting how isolated lab testing in a highly artificial environment has been hyped up as if it was a real thing. So maybe sometimes think about the way we test airplanes. Imagine if we
Speaker A
were to test an airplane in a wind tunnel and simulate a hurricane, and if in a wind tunnel I managed to create hurricane conditions to make an airplane break apart. But the newspaper then says, "Oh, airplanes can break apart
Speaker A
midair." Well, it never actually happened in real life. That, by the way, is what happened with the blackmailing episode. It turns out that you put AI in a lab and poke and probe it and have some of the most skilled
Speaker A
scientists in the world bend over backwards to try really hard to push the AI into the corner. So, it has no choice but, given all the conscious instructions, to kind of look like a blackmailer. That's what happened.
Speaker A
What happened in the media was, "Wow, AI is so dangerous. It could blackmail people." And the damaging thing about this is that a lot of these narratives have a tiny gem of truth in them. You know that yes, airplanes in hurricanes will
Speaker A
break apart in midair, but I don't know of many airplanes that broke apart in midair recently because we just don't fly them through hurricanes.
Speaker A
And so it's been very difficult to push back because you can't say airplanes will never break apart in midair. And you can't say there's no circumstance under which even a lab test you can't make AI do something bad. But
Speaker A
the amplification of these narratives has been very damaging for America because it's holding back what should be great opportunities for us to pursue, but instead, you know, we see kids turned off AI, people wanting to not use
Speaker A
these wonderful technologies that could make them so much better.
Speaker A
Yeah, I can speak to this. I'll tell two quick anecdotes. I don't know how many of you read the article in the New York Times about the AI companies doom trolling last weekend. And it seems on the outside of it kind of counterintuitive to a lot of people. Why would the AI companies be saying
Speaker A
they could cause the extinction of humanity? I think it was about eight months ago. I was part of a group of filmmakers that were invited up to Silicon Valley, and for a two-day symposium, and it was a lot of the bigwigs from DeepMind and Anthropic and OpenAI and all these various
Speaker A
engineers gave talks on the promise and peril of AI, but a lot of it was very scary, and they were saying, oh, within four to six years, four billion people are going to be out of a job. Well,
Speaker A
they took a hand count, and I can't name some of the people that were there, but there were people that, you know, in the AI pantheon, how many people think there's a chance that AI will cause the extinction
Speaker A
of the human race, and a third of these people raised their hands. You know, that there's a possibility.
Speaker A
Anyway, I'm not pushing back on what you're saying. What I'm saying is I experienced firsthand this. I was trying to figure out why the hell are they trying to scare us? It seems counterintuitive. And so anyway, you had
Speaker A
you have a lot to say. By the way, you have a very informative newsletter.
Speaker A
I think it's called The Batch, which I encourage everyone to join. It's super informative. Also,
Speaker A
I'm curious, you've come away with, I think, a sensible view of the odds of AI human extinction and all that. Do you want to share your views as well? Well, I'm kind of in the middle in terms of I think there are massive benefits. I think personally now I'm just a guy who does
Speaker A
vampire movies and stuff like that, so what do I know? But I'm in the middle. I think that the possibilities of AI superintelligence are probably farther out than people think. I think the possibility of AI—this is what I now
Speaker A
think—causing the extinction of the human race anytime within our lifetime is highly unlikely. I personally think, and we can talk about this in a few minutes, that the AI displacing workers, the threat might be bigger than you're saying it is, but
Speaker A
that's okay. We can talk, we can talk about that. And I've certainly seen that, experienced that in Hollywood to a certain extent. Um, so I'm in the middle, but I've I've also been using AI myself for the last two or three years.
Speaker A
And I've seen lots of ways that it has sped up my work, whether it be through um concept art iteration on something I'm doing or whether it be through visual effects or there are different ways that I use it in the writing
Speaker A
process and the research process as well. So, you know, it's it's a tool and like any tool, I think, you know, a hammer can be used to build and a hammer can be used to hurt someone. So, it it
Speaker A
just depends. You know, on on on the job issue that that you mentioned, um, one of the hype narratives has been AI will lead to millions of people unemployed and um, it does help people convince be convinced maybe that the technology is very
Speaker A
valuable, but I'm not seeing it happen for this reason. So, um, chatting with C of a large workforce management company that actually ran surveys of people that were recently laid off and we see all this news about AI leading to layoffs,
Speaker A
right? Company comes saying AI is coming, laying off people, blah blah blah. He ran a survey and asked of the people that were actually laid off, were you replaced by AI or not? And the number, what do you think it was? Was it
Speaker A
50%, 20%, 10%? It was 1.4% of recently laid off people actually had their jobs replaced by AI. To be clear, I'm very sympathetic to this 1.4%. I'm very worried. A tiny sliver, a small fraction of people have had their jobs replaced
Speaker A
by AI and it sucks for them and we need to do something about that. Give them a safe work with them safety net provide them a way to reskill, upskill, learn, you know, other things to to find new
Speaker A
jobs. So that's really important to hope this 1.4%. And again, it's one of these things where there's a gem of truth.
Speaker A
There's a tiny number of people whose jobs are going away because of AI, but the hype is vastly greater than the reality. And the broader reality is that um if you look at the taskbased analysis of jobs, these are very rigorous studies
Speaker A
that economists will do where we look at the jobs done by say the American economy or the global economy, break down the job into discrete tasks and then figure out what can AI actually automate. It turns out that for many
Speaker A
jobs AI could plausibly automate you know 30 40% of the task in someone's work. And it does mean that someone that doesn't use AI at all, they will be much less productive than someone that does.
Speaker A
And so that pathy saying that you people won't be replaced by AI, but people will be replaced by someone else that uses AI. That is absolutely true. But at the same time, if AI automates, you know, 30 40% of someone's job, you really need a
Speaker A
human to do that remaining 60 70%. Historical trends have shown that when this happens, when technology automates part of someone's job, the complement to it, the other thing that needs to go with it becomes even more valuable. And
Speaker A
so historically, the trends have been automation have led to higher wages for people. Which is why when we went from horses to cars, they became tons more jobs for taxi drivers because taxi drivers are a compliment for, you know,
Speaker A
the taxi for the car. Um, but slowing down people's willingness to adopt this technology is really hampering people's uh trajectory to get to these often better, higher paying jobs.
Speaker A
Um, okay, switching gears for a second. I'm curious, you know, you've briefed lots of policy makers on AI.
Speaker A
Uh so I'm curious to what degree do you feel our our government um to the extent that you can talk about it is sufficiently up to speed on AI and um what would you say is the the biggest
Speaker A
mischaracterization or blind spot that you've run into when talking about when talking to our government?
Speaker A
You know, I feel like in the US, the federal government has mostly got its head together in a in a more reasonable than more reasonable way than one would have hoped a couple years ago.
Speaker A
I was surprised to hear this. I see. Yeah. So, I mean, clearly we have a lot to drown to cover, but two years ago, the AI and extinction narratives were making us rounds in Washington DC. Um, and these were part of the regry capture
Speaker A
attempts of a handful of AI companies that want to say AI will lead to human extinction regulated. And in particular, it turns out that one of the major threats to some leading American AI companies is when you spend tens of
Speaker A
billions of dollars uh training large AI models called foundation models. Um, it turns out that some companies, including overseas ones in in in China and elsewhere, are training models and releasing them on the internet free for anyone to use. We call these openw
Speaker A
weight models. But it's really annoying if you spend billions of dollars to train an AI model and then someone else posts an equivalent one on the internet free for anyone to use. It really degrades the value of your of your
Speaker A
investment. So about two years ago, I was I I was actually kind of really surprised by the intensity of the lobbying to drum up fears on AI to try to pass stifling regulations to make sure that, you know, to make it
Speaker A
difficult for anyone to publish their AI models on the internet free for anyone to use. Fortunately, um, a lot of the AI community has largely beaten these these these efforts back at the federal level, but uh, the lobbying because there are,
Speaker A
you know, billions or more likely trillions of dollars at stake, you can imagine the resources and the intensity of the lobbying is intense. Um, there are also very credible reports that some of Americans adversaries are busy funding nonprofits to stoke up
Speaker A
unnecessary fears over AI safety to try to slow down American innovation in AI. The credible reports about our adversaries funding American and and other European organizations to slow us down is also very concerning. Having said that, I think uh uh both um white
Speaker A
house and congress have had enough talks with AI. You know, Schumer ran a bipartisan AI inside forum. Bunch of people including me spoke with Senate Congress. Um and I think is you know could have been much worse. I worry much
Speaker A
more about state level regulation of AI because a lot of the lobbying efforts have been now at the state level because you can get a state level stifling regulation passed. it could have deleterous effects across the entire nation and there's so many state houses.
Speaker A
So constantly squashing back referee proposals to the state level has been has been much harder.
Speaker A
So one of the one of the narratives that we hear in terms of you know not not not stifling innovation through overregulation is that at least in America uh that we're in a race with China. That's one of the things that we
Speaker A
often hear and and that if we don't get there first, China will become ascendant and then whatever the the West will be over. Um I'm curious uh because you're someone who's worked in AI initiatives for both the US and for China. So, one
Speaker A
of the questions that I asked you in kind of our pregame was how would you define the different approaches between the two countries in terms of AI, you know? So, um, one thing that I would not have guessed few years ago was that
Speaker A
China would end up being much more open in the way it innovates in AI compared to the United States. So again, I I think I I know I know the US ecosystem really well since I you know, both
Speaker A
Google Brain and Sam and I I might be the only person on the planet that both Sam and Dario has worked for, but so I I feel like I know all the teams of people well. Um but what has h at the moment
Speaker A
that CH GP was released uh in the end of 2022, America was decisively ahead of China. But China since then has played its hand really well leaning into the open-source openweight ecosystem in which teams in China will do research
Speaker A
and publish their research papers free on the internet for anyone to read. And by the way, I'll tell you a lot of the all the American teams read tons of the Chinese research papers and learn from them and benefit from them. Tons of the
Speaker A
techniques used by leading American teams now frankly were from you know re research innovations learned from China.
Speaker A
So I think there's very much two-way sharing of of of knowledge. Um but one thing that China did really well was a lot of its companies will um release this technology as open weight models meaning they're published on the
Speaker A
internet free for anyone to use including American companies but also can you just define what open weights are for the audience [snorts] how many of you know what open weights are some.
Speaker A
Okay, cool. Not bad. Oh, thank thanks for good good call. Thank you. So, when you have a AI model, read tons of text on the internet, look at tons of images, um, watch lots of videos. I know copyright is the sole point, but after
Speaker A
AI has learned from lots of data, it results in a long list of numbers. We call those numbers weights. And an open weight model is when uh someone has trained an AI model on lots of data and it publish and you publish all of those
Speaker A
numbers called weights on the internet free for anyone to download and use. And so when there's an open weight model any of us I could download it you could download it and have AI to run on your own laptop as opposed to the proprietary
Speaker A
closed weight models such as the leading models from OpenAI anthropic uh Google Gemini. Those are closed weight models because none of us can see the numbers that AI had learned. We can only send our prompt to one of these companies and
Speaker A
have them use these weights or these numbers compute the output, compute the response and show us only the response.
Speaker A
One brilliant move that Chinese companies did was publishing open weight models. It turns out to help the whole world but you help yourself the most because it rapidly increases the rate of diffusion of knowledge within your companies or within your ecosystem. And
Speaker A
so because of factors like that over the last few years, China has raised ahead in AI capabilities. And right now the US closed weight models is ahead of the Chinese models. But China leads the world in openweight models. And what
Speaker A
this means is for a lot of nations that want an open alternative um they're adopting Chinese models. So adoption of Chinese models in the US is you know there there's a lot but in nations such as Africa um the Chinese deep sea and a
Speaker A
handful of other Chinese models are adopted very widely far more than say the American open way models which have really fallen behind and this is important for two reasons. But first, just as storytelling we've seen in Hollywood is an important source of soft
Speaker A
power, do we want stories told about the importance of liberty and democratic values or do you want stories told that reflect other nations values? It turns out that when someone asks a question of AI such as what happened in Tianaan
Speaker A
Square in 1989, well, whose model they are using will cause the model to give an answer that reflects that nation's values. So getting your stuff out there is a tremendous source of soft power influence. And then the second thing is
Speaker A
AI is a key part of the supply chain of how we build a lot of software products.
Speaker A
And if um more and more companies are building on really a Chinese dominant supply chain that has implications for you know America that that right I think will also weaken our ability to to change the way the technology goes and
Speaker A
just these past two weeks um something a couple very strange things happened in America. First, when Enthropic released it latest model, the Fable 5 model, it put in place restrictions to stop Fable from helping AI researchers in ways that
Speaker A
it thought could support anthropics competitors. Um, and then like a week or so later, a few days later, the US government um imposed export control restrictions on the fable model causing Anthropic to shut down access uh all around the world to all customers. But
Speaker A
what this showed to the whole world is that private American companies such as Enthropic can shut down can restrict how others can use frontier AI models such as no helping competitors. It also showed the whole world that America can
Speaker A
impose export controls to yank access to AI technology. And so what I've seen just over the last week is this has accelerated in many capitals around the world the urgency with which they feel like they need to secure their own
Speaker A
supply of AI that the US government and private American companies do not have the ability to you know turn off at a at at a given moment's notice frankly. Um, and this is also causing many nations to look at openw weight models because once
Speaker A
they they have all the numbers, no one can take it away from them. Um, and I think uh this I think will probably have a unfortunate effect on on America's soft power.
Speaker A
See, I just assumed that happened because Trump was annoyed at anthropic and [snorts] Yeah, [clears throat] there's that theory wanted to get back at them. Um but but to be clear, you're a proponent of open weights.
Speaker A
It turns out that um if AI ends up concentrated in the hands of a very small number of gatekeepers, probably everyone here will be worse off. I mean, but the the very small number of gatekeepers would be better off because
Speaker A
they charge a very high toe, but almost everyone would be worse off. Um, and maybe I sometimes make an analogy to the mobile phone ecosystem. Today, there are two gatekeepers, uh, Android and iOS.
Speaker A
And if there are certain innovations they don't want anyone to really try out, then no one's allowed to try them out. But it turns out to be really profitable to be one or two of the gatekeepers. So, I think it would be
Speaker A
unfortunate if AI as valuable as is ends up with a handful of gatekeepers. And to be clear, a lot of friends that work at these AI companies, I hope they do really well. I hope they have successful IPOs, you know, do very well
Speaker A
financially. I think all that is great. At the same time, I think there is a path for these businesses to do really well without them becoming gatekeepers in a way that stifles the innovation of everyone else that would like to do
Speaker A
things on top of this technology. So, is there anyone aside from you uh uh in America that is is leading the charge on the open weight approach?
Speaker A
So, it's a tough one. Um Nvidia has turned out to be a major producer of openweight models. The their models aren't quite state-of-the-art, a little bit behind, but Nvidia selling GPUs has a financial incentive to do this. Meta
Speaker A
used to be publishing open way models uh but recently switched directions to closed way models and then there's a number of kind of you know small nonprofits um academic university labs working on open way models uh um but
Speaker A
because of the resources needed to build AI models has been challenging to get that going and the current White House administration has been reluctant for various reasons some of which I think are sound but we can debate them to put
Speaker A
been you know enough resources to fund an open alternative to proprietary models. Uh some of the Middle Eastern countries I think you know UAE has had some sign open way initiatives. Uh India number of nations are trying. So I think
Speaker A
there's actually a lot of interest in this but how well it plays out I think um I don't know remains to be seen.
Speaker A
So China might save us in this in this regard. Um okay switching gears for a second. I have a couple more questions. It's just having China control the open waste supply chain would would be uh I don't know it
Speaker A
created a very different world than the one No I No I know I'm being being physic I know I know I know um I one of the things that I've been thinking about is uh sort of what I call
Speaker A
the Plato's cave analogy. So I don't know, you know, in terms of the way that the models are currently trained, they're effectively trained on data that's been digitized, right? Whether that be text or pictures or, you know, videos or whatnot, sound files, and um
Speaker A
you know, there's a a massive amount of of humanities works that a have yet to be digitized. Obviously a disproportionate amount of western or possibly you know Chinese works have been digitized but there's a ton that hasn't been digitized and we all know
Speaker A
also that the internet is kind of a distorted model of our world right um so one of the things that I've thought about and been worrying about and you said a couple of reassuring things about this is that is that the models are
Speaker A
being trained on an an inaccurate depiction of our reality, a distorted depiction of our reality. And so, you know, if you if you just go by the internet, everything's brain rot and every woman is like a 36D and every man is, you
Speaker A
know, is a silver fox and, you know, all all of the all of the ad AI ads that were fed, you know, when we're um on CNN or any other site. So um to what extent is that a genuine concern
Speaker A
that the models are are basically trained on a distorted view of our reality that it doesn't accurately depict our reality?
Speaker A
Yeah. So I feel like uh everything you said right absolutely agree it's distorted view of reality and I want to so it turns out that um the AI teams I give them a lot of credit for correcting a subset of the distortions. So, it
Speaker A
turns out that there's a lot of races, sexist, you know, discrepancy texts on the internet. Um, but because AI teams have paid attention to this and deliberately squashed those sources of bias as best as they can, it turns out
Speaker A
that AI models today are far less, you know, racist and sexist than the typical human being. And in fact, um, I personally have really few ideas how to make a, you know, sexist human less sexist. I just don't know how to do
Speaker A
that. But I know how to make sexless AI much less sexist. We have technologies for that. So maybe it's a good thing that AI is less biased than humans are on average. But that's still not good enough. And then for the places that the
Speaker A
AI teams are not paying attention because there just too many forms of distortions, um then the AI models do persist with this type of, you know, bias or distortion or what have you. And this to me is another maybe another
Speaker A
reason why um we need people for a long time. For the foreseeable future, all of us will know tons of stuff or have tons of context that our AI models do not have and will not have in our lifetimes.
Speaker A
And so I don't know. By the way, my team's been trying for years to get AI to replace me. And I I I I kind of wish it was that easy. AI just isn't good enough. Well, your avatar is still
Speaker A
hallucinating. Yeah. Yeah. Yeah. That's right. And and maybe one one specific symptom of that is um I don't know how often you use AI to help you brainstorm, right? If I'm you know often if AI, hey, help me
Speaker A
brainstorm a marketing slogan for this or what's a good project idea for this. And when I do that, I find that often, you know, if I get 10 ideas, maybe there'll be one or two good ideas, one that's okay, and then six or seven
Speaker A
absolutely terrible ideas. And I go, how on earth could AI have thought that that's even possibly doable? Like I would never say that. And this to me is a symptom that AI is missing tons of context that it thought that ridiculous
Speaker A
idea was even useful to propose. So what I see is for the foreseeable future, AI will humans, we will all have a massive context advantage. Meaning there'll be a ton of stuff that we know about what's important for a specific job, for a
Speaker A
specific task, for a specific script you'll be working on. the AI does not know which is one of the reasons why I'm not that worried about you know AI replacing us for a long time maybe until someday that I don't know we all carry
Speaker A
cameras and microphones and all of our lives are digitized and AI advances far beyond what it is today but I that just seems you know like a very unlikely and if it ever comes to pass a very distant
Speaker A
future I'm sure most of the people here certainly I have had had the experience of of you giving a model a prompt whether it's for an image or a video or text and it clearly the AI you know misinterpreted or you
Speaker A
know uh the model you know misunderstood it didn't give me the desired result and so I'll think about it h okay how do I change my prompt to get it it's not it's not really understanding what I'm asking
Speaker A
and fortunately for me maybe that happens a tenth of the time but there there are plenty of times where I'll just give up [laughter] where I'll just say I don't know how to I don't know how to get it to understand what it is that
Speaker A
I'm trying to ask it. But I I've actually do do you want to share your work though for how you use AI to to I mean I mean just just in terms of writing uh like one of the things that I
Speaker A
started to do early on was I might write a uh uh a four-page monologue for something and and I I labored over it for you know x amount of hours and it's a really good monologue but I know that
Speaker A
it's too long. Uh, so, um, I I started not worrying about how long it was and I would just take the monologue and say, "Okay, I need this to be 40% shorter or two pages versus, you know, four pages."
Speaker A
So, I want you to kick out uh, six different versions of that two-page monologue. Again, you know, trimming down my own words, keep the context, whatever it is, but give me variations.
Speaker A
And and I'll also say like, you know, lean one variation this way, one variation that way, and within a minute or two, it will kick out six different versions of a two-page monologue. And none of them are 100% perfect, but
Speaker A
they're they're close. And I'll mix and match some of the phrases, and what might have taken me an hour or two cutting down that monologue now has taken me five minutes. And so that's an example of something that I'll use all
Speaker A
the time. or I was recently writing a script where I had a uh a character get kidnapped and zip tied and taken into the uh cabin of uh a high-end commercial helicopter. And I knew that I wanted to
Speaker A
have this character there about 100 feet over the ocean escape from the cabin of the helicopter and dive to his safety.
Speaker A
And so I said, "Okay, what are the highest end commercial helicopters out there?" yada yada yada. picked one and I said, "Oh, I want this guy to kind of maggyver a situation out of it." So, what are all of the things that would
Speaker A
likely come standard within this cabin that I can use for him to escape, whether it's, you know, the headphones, the wire from the headphones, um, etc., etc., etc., the the safety belts.
Speaker A
And and so I I was able to generate this action sequence, a combination of of of me giving prompts to the AI, writing, and then reiterating, but I was able to do that in I don't know, an hour or something
Speaker A
like that. So I would say that for features, it might take me anywhere from depending on the level of research, 8 to 10 weeks to generate a draft of a feature. But now I'll use AI and I can
Speaker A
usually do it in I don't know four to five weeks or something like that. So it's definitely sped up my writing. It's not writing the scenes for me but it's it's iterating and and using my words in different ways. So I I I don't know.
Speaker A
I've I've I found it helpful. I know that you are also have a company a foundation that um I'm I'm paraphrasing here. um works on AI initiatives, ethical AI initiatives for lack of a better word. Would you say that's right?
Speaker A
Foundation well or the projects that you help sure that that you help finance. We take responsible AI seriously.
Speaker A
Yeah. Yeah. Responsible AI. It turns out that uh when in my own work in AI and the teams I lead in AI, we do come across ethical decisions occasionally, but probably much less often than most people think. Um and then when we come
Speaker A
across these decisions, you know, frankly, we do wrestle with them. Um we don't know that we're always making the right decision. And there's a number of cases that we actually honestly don't know what the right decision is. And we
Speaker A
do probably what you'd expect. We have internal debate, consult with a range of stakeholders that represent a range of diverse perspectives, synthesize, sit down, hash it out, and then just make a call and, you know, really hope like
Speaker A
crazy that we got it right. Um, uh, some of them are really crystal clear. At AI fund, um, we have come across business ideas where we sat down and said, you know what, this idea is going to make a
Speaker A
ton. We think this idea is very promising from a financial point of view, but we think it's going to make people worse off and those are easy. We kill those projects really easily. Um the only tough ones which is not that
Speaker A
many is we honestly can't tell makes some people better off maybe make some people worse off. So does it at net net make the world better off? Those we really struggle with. Um I I guess a a side question to that is I
Speaker A
you have two kids, right? Uh how old are they? Uh daughter is seven, son is five.
Speaker A
Okay. I and I have three. Um uh 12-year-old, 15y old, and 19-year-old. And my 15year-old is just violently opposed to any use of AI whatsoever. I mean, it's just is just ardently opposed to any use of AI. And I
Speaker A
I I think I'm AI agnostic. So I'll I'll get in these debates with him about um I mean I think it's hard to argue that there aren't positive benefits of AI in terms of uh the medical industry, uh you
Speaker A
know, climate initiatives, all all sorts of things. Um but it's and and there's a big move now. you know, we've all seen the clips of uh at commencement speeches of people, you know, booing uh AI executives or whatnot. And I and I'm not minimizing
Speaker A
that, but there's a there's a sense amongst young people today, a lot of young people today, that uh they're just going to be squeezed out of a job and that they're not going to have uh any viable options in the future. and you
Speaker A
know we're going to find ourselves in in Brave New World, you know, in 5 years.
Speaker A
So, how do I know you don't believe that's true? Uh so, what would you say to a young person today growing up um in terms of how they should think about AI and how they should get involved in
Speaker A
AI? You know, I want to share with you what AI is doing to the software industry because it is a harbinger of what I think will happen with other industries as well, which I think would be very and
Speaker A
and so AI has learned to write code really really well, right? So AI coding assistants can write tons of code really efficiently. Uh because a lot of AI teams, you know, built by people like us, software engineers, and so we built
Speaker A
tools to help ourselves. And it turns out that it feels to me this is the most exciting time ever to be in the software industry because AI can't do a 100% of a software developer's job. It could do a
Speaker A
lot of the boring bits and so it feels way more fun now than it used to be, you know, even two years ago to be building software and AI. Moreover, I am hiring as many highly AI enabled software
Speaker A
engineers as I can find. uh these developers that are highly AI enabled are far more productive than the last generation of developers and leading businesses including mine would love to hire as many of them as we can find but
Speaker A
we just can't find enough of them. So the tragedy is a lot of these AI narratives have been turning people away from AI. So we're just not producing enough highly skilled software developers that know how to use AI that
Speaker A
know AI engineering. as AI starts to accelerate its impact on other industries. I think this will be true for other industries as well that if young people learn to embrace these tools, they'll be far more productive and be able to do far more work than
Speaker A
people without these tools. And on average, with that 1.4% 4% stat that I gave earlier as one concerning and important exception. Um I think that salaries will trend up and there'll be more job opportunities and the ability to do more exciting work and the tragedy
Speaker A
that now to be clear I think we do have a few problems during the pandemic interest uh near zero interest rates massive government stimulus many businesses did overhire and we're still suffering from some of the overhang of that leading to some amount of layoffs
Speaker A
that's very concerning so it's led to a tougher job market for young people than is ideal. But I think AI gets blamed for more more things than it truly deserves.
Speaker A
Um and uh I find that I know we I taught at still teach at Stanford. I love higher education. We should support higher education. At the same time, I feel like higher education has not adapted fast enough to AI, which is why
Speaker A
a lot of universities are still training students not even for the jobs of 2026, but for the jobs of 2022.
Speaker A
And those jobs are changing. And in fact, we don't want universities to train students for the jobs of 2026. We should be training them for the jobs of 2028 or 2030 and beyond. And unfortunately, higher education's process for adapting curricular is just
Speaker A
very slow relative to the speed of change that we need to meet this moment.
Speaker A
So with those important caveats, uh we have work to do to help young students get onto more promising career path. But those jobs are out there and we just can't seem to find enough people to do these exciting growing set of jobs. And
Speaker A
in case what I'm saying rings hollow, um oh maybe I share one other thing. My team right now in my office, we have I don't know, we have we have a bunch of uh uh interns. So, we have one high
Speaker A
school student, five, six college students. They're interns, all high VA enabled, and they're fantastic. I wish I could find more interns like that. I've been hiring interns this summer, and I love my interns. I hope some of them
Speaker A
will come back for permanent jobs. But all the interns I hire tend to be on the highly AI enabled side. Whereas, if someone kind of rejects AI, it's harder that what they can do ends up being very different.
Speaker A
Okay, I think we have time for a couple of questions. So, yes, go for it. Who Who has a question?
Speaker A
Hello. Yes. Great. Mr. Guyer, I feel like we didn't really get to hear from you so much. You mentioned that you used AI in your research. I'm very curious to understand how you weight your research from AI. I
Speaker A
apologize for the compound question, but also like do you weight certain AI platforms better than others because of uh their weighted distribution?
Speaker A
Uh okay. So yeah, anytime I'm doing research, you have to check the work uh I find of almost any any am AI model.
Speaker A
And I I if I'm doing research, I tend to prefer the models that site the sources so that I can click through to the original links. But yeah, I I always double check and I also assume that if
Speaker A
I'm using research, and by the way, I I I also interview people. there. It's not the sole means of research that I use when I'm writing a script, but I just assume that there will be mistakes, you know, at some point in it and that I
Speaker A
will have to vet that when I'm in the later drafts of it. So, but yeah, it's a mistake to just assume that. I mean, obviously when I was typing in my own bio, it said I had pancreatic cancers.
Speaker A
So, uh, I I personally use about six different models. I mean, I don't think there's any different models are better at different things. So, um, and I'm I'm by no means an AI expert, but yeah, I I use about
Speaker A
six different models and and I'm I try to keep semi up to date on, you know, the different uh innovations that are happening or new models that are coming out. Hopefully that answered the question.
Speaker A
Thank you for sharing your wisdom, gentlemen. Um, my question comes from your description on the failed marketing campaign, the six things that didn't fit your ideas, but maybe two of them did.
Speaker A
Do you think AI essentially will just be better for the one to 100 scaling or uh convergence process not very much the zero to one for us human beings to use whether you're building an app or a company or a movie and that's the first
Speaker A
part of the question the second part would be you mentioned the phrase um AI enabled do you think we could get to a day where we could have phrases like someone is AI empowered as a thing we throw around in workplace or
Speaker A
professionally and for the first part also if you agree that AI are better for one to 100 only how do we find the zero to one inspiration as you know growing artists or entrepreneurs out there um let's see I think AI is wonderful and
Speaker A
AI sucks for parts of the zero to one and for parts of the one to 100 um maybe from the zero to one there's a small set decisions are so consequential that it feels like human intervention is absolutely critical to steer it for
Speaker A
those high level scaffolding questions but then I find that even when um for the one to 100 if I'm interpreting the term correctly filling in the details it at least for me it still messes up frequently and and in fact David's
Speaker A
example of kicking out six uh ways of rewriting the monologue which is actually very you know decently advanced AI usage pattern uh still mess up some details. So you then cut and paste from different parts of it. Um and then I
Speaker A
don't know it's good question to to see what's the right terminology to use. Is it AI enabled or AI empowered or in Silicon Valley does this a AI tilled terminology that I don't love? Uh but maybe we should try to come up with a
Speaker A
good term around this. I I like AI empowered. I I will say personally that um well I I I I would split it in when I'm giving prompts for uh images versus prompts for texts I I find when I'm
Speaker A
giving prompts for images personally that sometimes uh you know it will generate something so bizarre that in a million years I never would have thought of that that we were using uh Midjourney when it first came out when we were on the second season of a
Speaker A
show I did called Foundation and we were using the my VFX supervisor and I were using it for early early iteration for concept design and then eventually we would come up with something interesting and then we would feed it to human
Speaker A
artists. But some of the times it would come up with [ __ ] that excuse my French was just so [ __ ] bizarre in in the but in the best possible way that we never in a million years ever would have
Speaker A
come up with that. And so I found when I was giving text to image prompting um the hallucinations were valuable because I would I mean like I'd have to be on acid all the time in order to come up with that kind of
Speaker A
stuff and and it was amazing and I would say holy [ __ ] this is amazing now human artists iterate on this but um but conversely when I'm when I'm prompting it for I' I've tried some experiments where it's like I'm
Speaker A
trying to come up with, oh, write me a a short film about X. Or if if you just give it the most basic prompt, it will so far it will tend to spit out the most cliched banal version possible. So you can't, but
Speaker A
people are a lot of people are lazy. And so I mean I I I have definitely read scripts that were generated by AI. You can you can you can you can tell you can't so I don't think at least in terms
Speaker A
of pros and I'm not talking about like writing a sonnet or something like that but like a script I don't think that the zero to one is there I think you've got to rely on your own human experience
Speaker A
personally anybody else hi there my name is Ted Sim um this question is for Andrew uh you know as a thought leader in AI There's so many industries out there and in particular, we're talking about Hollywood here. So,
Speaker A
I want to ask about your particular interest in Hollywood and the conflict right now between Hollywood and AI. Do you see any urgency for Hollywood to need to lean in and become more AI enabled? And if so, can you paint that
Speaker A
picture of what that could look like if we don't participate? Yeah, thank you. I really got your stats. Um, one of the I wish Hollywood would tell more positive AI stories. Um, I feel like Hollywood, you know, many people in
Speaker A
Hollywood have been worried about job displacement from AI. Um, I think 1.4%, right? I don't know what the number in Hollywood will be. I don't say don't worry about it at all, but it's not going to be nearly as bad as most people
Speaker A
fear. But for various reasons including the job worries, we see a sequence of very negative stories about AI coming out of Hollywood. And we know that the stories we tell influences culture.
Speaker A
So what we're built in America is a country where the vast majority of our kids and even many grown-ups do not like AI and do not want to embrace it and want to put a stop to it and do not want
Speaker A
to build data centers. And this is very damaging for our nation. In comparison, in China, you hear a lot of stories being told about some young entrepreneur that worked really hard and built a um solar solar farm and used that to power
Speaker A
data centers and because they did so well, they met the true love and guess what the young kids therefore are motivated to do. I was really moved when one of my teams working in K12 education uh met this high school student and she
Speaker A
said I heard AI has something to do with human extinction and I don't want anything to do with that and so she self- selected out of a career that had anything to do with AI at this moment when AI is the most promising it has
Speaker A
ever been. So she made a terrible I think really terrible career move or maybe not a great career move. And so if Hollywood in addition to all the you know Terminator scary AI stories we tell can give a more balanced view that AI
Speaker A
could also solve a lot of problems. I think that that would shape the national narrative in AI in a much more healthy way and it is urgent.
Speaker A
Hi. Uh, my name is Michael Lander. I'm a writer, director. I use AI very similar to you. Um, I have found in my sort of creative circles and being an impath of feel like when I say I use AI, I kind of
Speaker A
get like your son, your 15-year-old. Um, you know, it's AI is built on stolen property. And uh there are times where I have my own sort of answer to those folks who say that to me. I would love
Speaker A
to hear sort of the technical answer to that and also the artist answer to that.
Speaker A
I'll do the artist answer and you could do the technical answer. It'd be funny if we swapped. Uh um it's funny. My I mean my son is like he's like I [ __ ] hate that you do that. Uh um
Speaker A
[sighs] I think that to first of all I think AI the output of AI is only as good as the input of AI. Okay. So I I know that I personally and I for those of you who are in the creative community here, I've
Speaker A
definitely received notes from development executives that were generated by AI. I mean 100%. And and the notes are terrible because some it probably isn't even the executives and an intern or an assistant just typed in how do we make this script better
Speaker A
period, right? and it sped out the most generic set of [ __ ] notes, you know, that you could get. And and in some cases, I've been given notes where there were they obviously they didn't even read them before they gave them to us,
Speaker A
right? Because there were all sorts of, you know, mistakes in them. So, it's only that that is is key is it's only as good as the input, right?
Speaker A
uh if if if we as artists are just giving it prompts and saying make me a hit song or make me a hit movie and that those are the that's the only intervention that we're doing, it's going to create crap. It's going to
Speaker A
create generic crap. But if we use it as a sounding board or if we say, "Okay, I've written this scene. Now give me 10 really surprising ways that the character could try to back out of that date or something like that." Again, it
Speaker A
will give seven of them probably won't be very good, but in my case, maybe three of them will be interesting or interesting enough that it'll take me down a path that I hadn't thought of before. So, I think that again, it's a
Speaker A
tool. I mean, I know that when we were switching from film to, you know, digital cameras, so many I've worked with Chris Nolan a lot. He absolutely refuses to do it. Gone Del Toro embraced it immediately. Uh I now that I've switched to digital that
Speaker A
I I would never go back to film because there's so many more uses for it. Uh and so I I regard AI as a tool in that regard and we've all seen terrible uh digitally shot films and we've seen
Speaker A
amazing digitally shot films. And so I again I just think it's a tool and we just can't be lazy about its use. You know, it's it's got that the human has to constantly I we've heard this phrase human in the loop where uh but there's
Speaker A
another phrase that I certainly didn't coin where I I just think it has to be human in the the lead and and you have to remain human in the lead. You can't you can't just outsource your creativity if you want to create anything
Speaker A
of any note. And I do think last point and then Andrew please take it from here. I do think we're quickly moving to a place whether it be songs on Spotify or short form content or eventually long form content where there's really going
Speaker A
to be two tiers of of art. There's going to be the stuff that's basically just barely human in the loop and then there's going to be stuff that's human in the lead and but the human elite is very
Speaker A
interesting uh concept right somehow AI has learned from the internet and the AI technology tends to average all the stuff is red um which is why we end up with this middling you know slop or milling average rather than something
Speaker A
extraordinary. uh unless we push the technologies in certain ways and I feel like as a technologist I know the technology well but I'm not nearly you know the storyteller that right almost all that that all of you are so I should love to
Speaker A
find ways as a technologist to do more projects with Hollywood I'm actually working on one that I'm not allowed to talk about yet with a Hollywood team but it'd be great if there's a way to do more projects with all of you to ideally
Speaker A
tell more positive stories about AI I'm gonna come up to Silicon Valley and we'll Oh yes, we'll talk about that. Yeah, we should do that deal.
Speaker A
Ju just to answer the AI built on stolen uh works remark. I I want to share my personal view. Um you know earlier today I was at the broke museum ponder um who's also here and uh and I love the
Speaker A
Getty Museum and when I visit you know some of these museums um uh sometimes I see the young artists sitting on the floor um using the paint brushes to copy the work of the great masters right not at the bro but maybe at the Getty or the
Speaker A
Smithsonian and so on. Um, and so we as a society have decided that we like it when young artists sit on the floor to copy these century old paintings as part of how they master their craft. And they're not allowed to kind of repaint
Speaker A
someone's painting verbatim and pass it on as an original. They're certain barriers to that. But we encourage our young artists to learn from the great masters.
Speaker A
Today we also have young authors that will read tons of books, read things on the internet and synthesize that to form their own creative style of writing or their own creative style of painting.
Speaker A
And if we allow young artists to do this, um I would quite like as a non-art or as some to be allowed to send AI to do this for me. I think there was a line that should not be crossed. Uh there was
Speaker A
a lawsuit against Anthropic where Anthropic was found uh to have used copyrighted books without permission including three of mine.
Speaker A
Yes. So that was part part of that lawsuit. So I think that did cross a certain line but well well they crossed the line because they they knowingly uh took a trove of of books that were um digitized without permission. and they
Speaker A
knew they knew they were doing that. So yeah, I it was absolutely crossing the line. But sorry, keep going.
Speaker A
Which fair point I actually I actually think you know using copyrighted works that way I think is problematic. And then at the same time um if I'm allowed to read a web page that someone chose to freely publish on the internet and I'm
Speaker A
not allowed to copy a verbatim that would violate copyright. But I'm allowed to read tons of web pages and learn from that and take inspiration to synthesize my own view to write a brand new article that doesn't copy anything verbatim. It
Speaker A
feels okay to me to be to let to say that instead of me doing it personally, maybe I can send AI to do it for me. Um at the same time because if we think humans can do this, can we let humans
Speaker A
send AI to do the same thing for them without verbatim copying? Now I realize that there's an important compensation and fairness question that needs to be answered too because AI can now do this at a scale that no human can which seems
Speaker A
like a positive thing but um longer term how do we figure out an economic model to make sure that it all still works. Um I think it'll actually work out okay but but I I I realize that there's a lot of
Speaker A
angst for the publishers that worry about the financial models as well. Okay. I wish we had time for more, but I think yes, unfortunately, I think we're all out of time, but can we hear it again for our incredible speakers? [applause]
Speaker A
Thank you both so much.
Topics:Andrew NgDavid S. GoyerArtificial IntelligenceAI hallucinationAI risksAI creativityAI regulationHollywood AIAI job displacementAI narratives
