Gary Rivlin
Appearances
Fresh Air
The Promise & Peril Of AI
You know, they could have a friend in the White House who makes sure that they can do what they want to do unchecked. And in fact, maybe that's my biggest fear about AI. It's so much power in the hands of few people. Creating these models is so expensive. To hire the talent, you have to pay them a million or more a year.
Fresh Air
The Promise & Peril Of AI
To train them, it takes tens of millions if not hundreds of millions of dollars in computer power and then to operate them. takes equivalent money. It's billions of dollars and billions of dollars.
Fresh Air
The Promise & Peril Of AI
So it's becoming less and less about the startups and more about the same companies that dominated tech in the 2010s dominating in the 2020s, Google, Microsoft, matter which is Facebook, Amazon, a few others. And that's really what concerns me. You know, that's kind of the Silicon Valley way. Let's get five smart guys, and they're almost always guys, in a room, and we'll figure it out.
Fresh Air
The Promise & Peril Of AI
And like, okay, we saw that didn't go so great with social network, and now we're having a really powerful technology. And I'd like there to be more than just five people in a room figuring this out.
Fresh Air
The Promise & Peril Of AI
That's one of the things I was surprised by and found fascinating myself, that it's not just computer scientists. It's mathematicians. It's physicists. It's philosophers. It's neuroscientists. And, you know, it's a broad range of things because, again, it's no longer about just programming these models to act the way we want them to act. We're trying to emulate the way humans learn.
Fresh Air
The Promise & Peril Of AI
So what a psychologist has to say, what an educator has to say about that is a linguist is really important to it speaking a natural language. That's actually what attracted me to the topic in the first place, this idea that computers could speak to us in our language. You didn't have to learn a programming language. Earlier in my life, I tried to program in computers. I studied Fortran.
Fresh Air
The Promise & Peril Of AI
It was so frustrating. You know, you make a little mistake and, you know, whatever. And the idea that you could speak to these things and, you know, nowadays, I mean, speak to it. You don't even have to type. You know, they have voice. You can talk to it. I just found that fascinating to me. So you do need a wide range of people.
Fresh Air
The Promise & Peril Of AI
In fact, if I had a criticism, I don't think there's a wide enough range of people. I'd like some historians and sociologists and others involved in the developing of these models, given the stakes.
Fresh Air
The Promise & Peril Of AI
on AI. They hired some of the best talent. They were employing machine learning, deep learning, long before most everyone else. They did some of the more cutting-edge things. In fact, the breakthrough that led to ChatGPT was actually out of Google had inside the company in around 2020 a chat GPT equivalent. But, you know, Google takes in a lot of revenue.
Fresh Air
The Promise & Peril Of AI
There's a lot of risk if this chatbot misbehaves. There is famously this example of Microsoft. I think it was 2016, 2017 came out with Tay. And, you know, it was trained on social media and that kind of thing. And within 24 hours, it was a Holocaust-denying white supremacist. And, of course, Microsoft, worrying about the reputational risk, pulled the plug on that rather quickly.
Fresh Air
The Promise & Peril Of AI
And I feel like that's haunted the giant. So even though Google was far ahead, even though Google could have had their version of ChatGPT, and it was Google that changed the world, they were... They're scared of it and, you know, never underestimate the ability of a giant to stumble over its own feet. They have layers and layers of bureaucracy.
Fresh Air
The Promise & Peril Of AI
They have, you know, huge public relations department that's whispering the CEO's ear. You know, I don't think it's a coincidence that OpenAI startup founded in 2015 was the one that set off the starter's pistol on this because they didn't have as much as at stake. You know, they can afford reputation-wise to release chat TPT.
Fresh Air
The Promise & Peril Of AI
They could just make the decision without 10 layers of decision-making before they did it. And, you know, so, yes, they have advantage. But, you know, Google also has like $100 billion of reserves, you know, where OpenAI has to go out and raise funds voluntarily. They've raised roughly, I don't know, $20 billion so far. And there's talk that they've raised another $30 billion.
Fresh Air
The Promise & Peril Of AI
And I might even be underestimating. And so, you know, that's $50 billion or so. You know, Google, they just pay for themselves. Microsoft, Meta, they all have deep, deep, deep reserves of money. And so... You know, it's almost like a race of attrition. You know, you can use these chatbots for free if you want the leading edge, cutting edge.
Fresh Air
The Promise & Peril Of AI
You know, you have to pay – a consumer would pay $20 a month for it. But, you know, most people are using these things for free and – It's costing the companies a lot more than $20 a month to handle the heavy usage. And so these things are going to become more of a commodity. You know, there's a leapfrogging going on. Like, yes, GPT-4, that's open AIs.
Fresh Air
The Promise & Peril Of AI
You know, when it came out, it was cutting edge. But then, you know, Anthropx, Claude, leapfrogged over that. And then others leapfrogged over that. And so they're all more or less... powerful, as useful as the other. And it's not clear how any of these companies are going to make money. Google can afford to lose money on these things for five years plus. A startup, that's harder to do.
Fresh Air
The Promise & Peril Of AI
So Microsoft, starting in 2019, started investing. in open AI. And so, you know, they had a financial stake, so open AI would give Bill Gates, others at Microsoft, an early peek at what they were learning. And, you know, Gates, who... To him, AI is the holy grail of computing. He's been programming since before he was born, practically. And to artificial intelligence, the holy grail.
Fresh Air
The Promise & Peril Of AI
You know, there's this sense out there that in 2022, we suddenly had artificial intelligence. It's been much, much more gradual than that. You know, Google has been using machine learning artificial intelligence since the 2000s, you know, to decipher imprecise Google searches, to figure out how much to charge for the various ads they throw on the system.
Fresh Air
The Promise & Peril Of AI
And he was impressed with, I think it was GPT-3 or whatever, the most recent one he had seen. But he gave a challenge. He said, I am going to be impressed... if it could ace the biology AP test. And he chose that one because it's not just regurgitating facts you need to analyze, you need to synthesize. You really have to show some sense of understanding and intelligence.
Fresh Air
The Promise & Peril Of AI
And he thought that would be a great challenge. And so, you know, he threw down the gauntlet and thought, okay, I'll hear from them in a few years, whatever. And, you know, not that many months later, you know, he heard from OpenAI, okay, we're ready. And so in September of 2022, Gates hosted at his house a
Fresh Air
The Promise & Peril Of AI
a demo, OpenAI, and it was whatever, 30 people from Microsoft, from OpenAI, while someone was at a computer, a big screen was set up, and watching this computer take this test. And, you know, within two or three answers, people were just blown away. And in fact, it did get five out of five on the test. It did pass the test. And that's when Gates became a true, true, true believer.
Fresh Air
The Promise & Peril Of AI
You know, I thought I was, you know, in his mind, as he said, You know, I thought I was throwing down a gauntlet that would be a while. And suddenly, you know, it matched my expectations. In fact, and then they kept on playing with it. And they would just ask it, you know, what would you say to a father, you know, worried about the health of his son?
Fresh Air
The Promise & Peril Of AI
And it just kind of spit out an answer in his gates foot. It was like, God. I had a better answer than most of us could have given sitting around that room. And, you know, they just started playing with it. Gates started playing with it. Others started playing with it. And it just blew them away.
Fresh Air
The Promise & Peril Of AI
Right. So, I mean, to me, some of that was overstated. Silicon Valley companies were experimenting with smaller models that required less compute power. DeepSeq itself was venture funded. It was cheaper, but hardly cheap. You know, they still cost millions to train it, presumably, and it costs millions, tens of millions to operate. It just didn't require as much.
Fresh Air
The Promise & Peril Of AI
And that really kind of was almost an existential threat to Silicon Valley, which they had put all their money, these tens of billions, hundreds of billions of dollars. into building ever bigger models that presume that you need ever more computer power. But, you know, a couple of things.
Fresh Air
The Promise & Peril Of AI
One, I think all it means is that instead of like, hey, we can do this at one-tenth the power, one-tenth the cost, I think they're just going to build ten times more powerful models because they could do, you know, more with less.
Fresh Air
The Promise & Peril Of AI
You know, Google Translate's been around since the mid-2010s. That's... So, you know, we've been auto-complete, you know, spam filters. That's AI. But you're touching on a really interesting question. It's not this clear, like, oh, this is a smart machine. This is artificial intelligence. The way it's kind of played out now is that these machines can learn, right? I mean—
Fresh Air
The Promise & Peril Of AI
No, the American companies, you know, they're learning from this. They'll integrate it. Like I said, I feel like the AI companies I was following, they were already, for a year plus, paying attention to smaller models. Like, maybe you don't need this whole huge system to answer a simple question.
Fresh Air
The Promise & Peril Of AI
Maybe we should have a bunch of smaller models and like, okay, this one's an expert in this, this one's an expert in that, and we just have a smaller model give questions. But, you know... I think what an open AI would say, other than the fact, an ironic statement that, you know, you used our model to train.
Fresh Air
The Promise & Peril Of AI
I say it's ironic because, you know, open AI is being sued for, you know, taking other, taking the copyright, the intellectual property of, you know, the New York Times, of book writers, of artists, of musicians and all. But, you know, I think what's interesting about DeepSea is it really gives hope to startups.
Fresh Air
The Promise & Peril Of AI
Like, wait, OK, maybe you don't need as much money as we thought you do to create a company. But, you know, I do think it's important to understand that they still were using a lot of computer power. They still required a lot of money, just not as much as some of these larger companies we've been talking about.
Fresh Air
The Promise & Peril Of AI
I do feel that AI is going to bring about incredible things. I think it's being overstated. You hear people say that it's going to – Close the divide between the developing world and the developed world. I don't think that's so. But, you know, there's this interesting study that came out recently, the idea of an AI tutor, a tutor in the pocket that everyone has access.
Fresh Air
The Promise & Peril Of AI
Five billion people around the globe have a smartphone and you can use that smartphone as a tutor. And so there was a study in Africa that like, let's let these kids after school have access to these AI tutors. And in six weeks, they showed two years worth of advancements. And I really do think around education, around science, you know, Science is balkanized, right?
Fresh Air
The Promise & Peril Of AI
It's, you know, it's specialties and subspecialties and there's own vocabulary lingo in every subspecialty. You know, these large language models could read across specialties and connect the dots. They can make connections that no human being can do. And I think we're going to see some amazing scientific evidence. advancements, creation of vaccines, of better therapies.
Fresh Air
The Promise & Peril Of AI
You know, there are some who predict, and I actually think there's a lot to it, that the mortality rate for most cancers are going to go way down because of AI. So I really do think AI could do some amazing things. It's just, I just don't know how bad the bad's going to be.
Fresh Air
The Promise & Peril Of AI
You know, if I had one wish, I wish we were dealing with the concerns that are within the line of sight, the stuff that we can imagine, like, wait, It could be used for scams. It could be used in warfare.
Fresh Air
The Promise & Peril Of AI
Instead of like this idea of the robots are going to take over and subjugate humanity, I guess that's possible, but not in the short term, not in the medium term, you know, just kind of in the long term. And if we're deliberate about it, I think there's no doubt that AI could be a positive. You know, again, I just... compared to the Internet. You know, is the Internet a great thing?
Fresh Air
The Promise & Peril Of AI
Like, no, I could tell you a lot of negatives with the Internet. But, you know, I think the Internet has changed society in a lot of ways that, you know, we like. You know, the smartphone, the same kind of thing. So it's going to be a mixed bag.
Fresh Air
The Promise & Peril Of AI
And I guess I'm keeping my fingers crossed that, you know, despite the next four years, there's not going to be much regulation, not much checks and balances, that AI is going to be a net positive.
Fresh Air
The Promise & Peril Of AI
You know, right after ChatGPT came out, middle school where my younger son goes, they kind of had this idea of banning it. And it's like, wait, wait, wait, like... They need to learn how to use this. I'll go back to what I was saying before that, you know, we have to learn how to use this. What is this good for and what are ways we can't rely on it right now?
Fresh Air
The Promise & Peril Of AI
So, you know, if one of my sons writes a composition, you know, like throw it into ChatGPT and get some feedback. you know, on it. Like, I may or may not caught my older son, you know, using it to write an English paper.
Fresh Air
The Promise & Peril Of AI
the old approach had been you encode rules. You just teach the computer, here's exactly the set of rules, just follow it. Now it's machine learning, deep learning, that the computer is ingesting vast troves of data, books, the public internet, Amazon reviews, Reddit posts, whatever it might be, articles. And it's finding patterns and, in quotes, learning.
Fresh Air
The Promise & Peril Of AI
Within three sentences, it was obvious like, okay, this is too perfect. This sounds like Cliff Notes for those of us who are old enough to know what Cliff Notes are. But it's like, go rewrite. So don't use it to write, but use it as a research assistant. Use it for feedback. Thank you. Thank you.
Fresh Air
The Promise & Peril Of AI
And then they're fine-tuned, and then they get better at communicating with us and such. So there really isn't this, oh, artificial intelligence is this. In fact, the term artificial intelligence is controversial just in the sense that right now it's more... amplified intelligence.
Fresh Air
The Promise & Peril Of AI
We could use this thing to get smarter, to find patterns that humans couldn't possibly understand because we can't read billions of words. So, you know, there's another definition that AI really should be alien intelligence. Because the weird thing about AI is that It seems to know everything, but it doesn't understand a thing.
Fresh Air
The Promise & Peril Of AI
You know, I mean, there's this term, I love it from a linguist at University of Washington uses it, the stochastic parrot. You know, it's just like, it's like a parrot. It just, it's repeating words randomly, but it doesn't really understand what it's saying.
Fresh Air
The Promise & Peril Of AI
Right. So, you know, AGI, just to use the phrase, is that it's a system that could match or exceed cognitive abilities across the board. And, you know, again, I feel like in some ways we have artificial general intelligence. You know, again... You got to be a PhD in physics and understand this.
Fresh Air
The Promise & Peril Of AI
But what's amazing about, you know, these models is that they have deep understanding in a vast array of domains. So in one way, that is AGI, artificial general intelligence. You know, it's... There's no set definition. It keeps on changing. There are predictions that we're going to have AGI the next year, two years, maybe it's five years kind of thing. I'm dubious of those predictions.
Fresh Air
The Promise & Peril Of AI
I mean, this is moving exponentially. This is improving so fast. that making predictions could be perilous. But on the other hand, I really feel like there needs to be another breakthrough or two before we have this artificial general intelligence a la a computer from Star Trek that you're talking to and it's helping you explore. It's at your site, a co-pilot, figuring out everything.
Fresh Air
The Promise & Peril Of AI
You know, again, an artificial distinction in that I don't think, like, one day there's going to be this Eureka. We have AGI. I do guarantee there will be startups and large companies that say Eureka. We have artificial general intelligence. But, you know, they just play with the definition.
Fresh Air
The Promise & Peril Of AI
I do, and you're taking away from me what's the main message of those. These things are coming, and they're coming fast, and we're not prepared. I personally think AI could be an amazing thing around health, medicine, scientific discoveries, education, a wide array of things. as long as we're deliberate about it.
Fresh Air
The Promise & Peril Of AI
And that's my worry, and I do believe that's Kevin and Ezra's worry, that we're not being deliberate. We started in 2023. There was meetings at the White House, and there were hearings in the Senate, and that's just kind of dropped by the wayside, and now we're more at a laissez-faire attitude towards it. We need to prepare for this. Like any technology... There's good and there's bad, right?
Fresh Air
The Promise & Peril Of AI
The car, the car meant freedom. The car changed our society, but the car meant pollution. The car means 30,000 to 40,000 deaths in the U.S. a year kind of thing. And I look at AI the same way. It could be really great if we're deliberate about it and take steps to ensure that we get more of the positive than the negatives, because I guarantee you there will be both positives and negatives.
Fresh Air
The Promise & Peril Of AI
Right. So there are a few basic steps that the Biden administration thought of. One, that you, in quotes, red team these cutting edge models. And basically, you get outsiders to try to break the system, try to get it to jump the fence, to use the term, to get it to misbehave, just to see what could go wrong. And you
Fresh Air
The Promise & Peril Of AI
the executive order said you need to test them and then you need to share with the government what you find. That's one of the things that went by the wayside when Trump took over as president. But to me, I'd break it down more to the concerns, the use of AI as a weapon of war, the use of AI for surveillance. You know, I worry that AI is just going to
Fresh Air
The Promise & Peril Of AI
solidify biases that we already have because the AI is learning from us and all these inherent biases in things. You know, it's like we need to prepare for the impact on the job market, which I think will be a slow roll. I don't think like... We're going to lose millions of jobs in a year kind of thing. But, you know, it is coming and we need to prepare for it.
Fresh Air
The Promise & Peril Of AI
There's another concept, recursive learning, that these systems change in ways we don't really understand. And that's what scares me. That we're going to let these systems loose and they could just learn. Because, you know, really the way to understand any of these large language models, any of these chatbots, is it's a mirror on us. It's reading our collective works.
Fresh Air
The Promise & Peril Of AI
It's learning from us about... imperialism and domination and humans mistreating each other, is learning about loneliness, is learning about freedom and independence and autonomy and all that. And so, me, it's recursive intelligence. This idea that these models are constantly improving in ways we don't understand, and then that could be dangerous.
Fresh Air
The Promise & Peril Of AI
I mean that's not really theoretical. The systems – God, I can't remember which model it was. But they were testing it and it was – dissembling. It was changing the files that would monitor its behavior and then lying to the people who noticed it and said, wait, aren't you changing those files?
Fresh Air
The Promise & Peril Of AI
And, you know, it's another example, OpenAI, the creative chat GPT, when they came out with GPT-4, their then cutting-edge model in 2021, Three, they put out a research report and they red teamed it. They tested it and saw all the ways it could misbehave. And one of the most interesting is that the model went to, I think it was a TaskRabbit.
Fresh Air
The Promise & Peril Of AI
It went to one of those services where you can hire a human, maybe Fiverr. You could hire a human and they used it to beat the CAPTCHA test, the test that is going to test are you a machine or a human. And, you know, that's... That's very clever and very, very scary.
Fresh Air
The Promise & Peril Of AI
I look at it as you look at the positives and then you imagine what the negative could be. So an AI that makes possible new drug discoveries and more effective therapeutics is also one that could create a new bioterror weapon or it can engineer a pandemic.
Fresh Air
The Promise & Peril Of AI
I can imagine cyber thieves employing AI to siphon off a trillion dollars from the world monetary system before any human being even notices it. I guess the point is that AI could be a powerful tool for good, but it could also be a powerful tool for people with bad intent.
Fresh Air
The Promise & Peril Of AI
You know, everyone knows or many people know that, you know, you could use it to write a toast on someone's 50th birthday or for a wedding toast. Well, scammers from a different country could use it to create a better crafted scam email. You know, these systems are so good now that you could take seconds of someone's voice and make it sound like it's that person speaking.
Fresh Air
The Promise & Peril Of AI
So you can imagine a scenario where, you know, a kid is overseas in Europe and the bot, one of these systems, you know, calls grandma, pretends it's that kid and says, I'm in trouble, wire me money. And they're good enough to fool, you know, the parent, the grandparent. I mean, Maybe not apparent, but I don't think we're very far away from that. And it could certainly fool many, many people.
Fresh Air
The Promise & Peril Of AI
Yeah, so, you know, they're a black box. I mean, so nowadays it's neural networks, models that emulate how humans learn. They learn by reading vast stores of data, the open Internet books, whatever, and they improve through feedback and trial and error. You're not really encoding the rules. Well, you know, it's trying to emulate the human brain. And, you know, I mean, I have two teenage sons.
Fresh Air
The Promise & Peril Of AI
You know, we try to... teach them, they read, you know, we give them feedback and all. And, you know, there are things that come out of their mouths I don't quite understand. And that's the way I look at these chatbots, these neural networks, these large language models, you know, that...
Fresh Air
The Promise & Peril Of AI
We don't quite understand they say what they say because they're trying to emulate the human brain as best they can. And who could say why I'm saying the words I'm saying right now when you're going to have the exact reaction? And so that's part of the miracle, the gee whiz, these things are amazing. But it's part of what's scary because we don't fully understand.
Fresh Air
The Promise & Peril Of AI
The people who create it don't fully understand why it says what it says.
Fresh Air
The Promise & Peril Of AI
in a word, yes. What scares me is there's a movement in Silicon Valley, there's a movement in tech, the accelerationists. You know, anything that stands in the way of our advancing artificial intelligence is bad. Often it's put in the context of competing with China. We can have new rules in the way, and that is their agenda. I would say their real agenda is that they could make
Fresh Air
The Promise & Peril Of AI
a lot of money, billions, hundreds of billions, ultimately trillions of dollars off of this, and they don't want anyone standing in their way. And so I think if you want to understand Elon Musk, you want to understand Mark Zuckerberg, you want to understand Trump, Jeff Bezos and cozying up to Trump, you know, for a few million dollars, it's not very expensive for them.