We started seven years ago because we felt like something really interesting was happening AI and we wanted to help steer it in positive direction. It’s honestly just really amazing to see far this whole field has come since then. And it’s gratifying to hear from people like Raymond who are the technology we are building, and others, for so wonderful things. We hear from people who are excited, hear from people who are concerned, we hear from who feel both those emotions at once. And honestly, that’s how we feel. all, it feels like we’re entering an historic period now where we as a world are going to a technology that will be so important for our society going forward. And believe that we can manage this for good.
So today, I want to show you the current state of technology and some of the underlying design principles that we hold dear.
So the first I’m going to show you is what it’s like build a tool for an AI rather than building for a human. So we have a new DALL-E model, generates images, and we are exposing it as an app for ChatGPT to use on behalf. And you can do things like ask, you know, suggest a nice post-TED meal draw a picture of it.
(Laughter)
Now you get of the, sort of, ideation and creative back-and-forth and taking care of the details for that you get out of ChatGPT. And here we go, it’s not just the idea for the meal, but very, very detailed spread. So let’s see what we’re going to get. But ChatGPT doesn’t just generate images this case — sorry, it doesn’t generate text, it also generates an image. And that is that really expands the power of what it can do on your behalf in terms of carrying your intent. And I’ll point out, this is all a live demo. This is all generated the AI as we speak. So I actually don’t even what we’re going to see. This looks wonderful.
(Applause)
I’m getting hungry looking at it.
Now we’ve extended ChatGPT with other tools too, example, memory. You can say “save this for later.” And the interesting thing about these is they’re very inspectable. So you get this little pop up that says “use the DALL-E app.” And by the way, is coming to you, all ChatGPT users, over upcoming months. And you can look under hood and see that what it actually did was a prompt just like a human could. And so sort of have this ability to inspect how the machine using these tools, which allows us to provide feedback them.
Now it’s saved for later, and let me show you it’s like to use that information and to integrate other applications too. You can say, “Now make a shopping for the tasty thing I was suggesting earlier.” And make a little tricky for the AI. “And tweet it out all the TED viewers out there.”
(Laughter)
So if you do this wonderful, wonderful meal, I definitely want to know how it tastes.
But you can that ChatGPT is selecting all these different tools without me having to it explicitly which ones to use in any situation. And this, I think, a new way of thinking about the user interface. Like, we so used to thinking of, well, we have these apps, we between them, we copy/paste between them, and usually it’s a great experience within an app long as you kind of know the menus and know the options. Yes, I would like you to. Yes, please. Always to be polite.
(Laughter)
And by having this unified language interface on top of tools, AI is able to sort of take away all details from you. So you don’t have to be the one who spells out single sort of little piece of what’s supposed to happen.
And as I said, this is a demo, so sometimes the unexpected will happen to us. But let’s take a look the Instacart shopping list while we’re at it. And you can see we a list of ingredients to Instacart. Here’s everything you need. And the that’s really interesting is that the traditional UI is very valuable, right? If you look at this, you still can through it and sort of modify the actual quantities. And that’s something that I think shows they’re not going away, traditional UIs. It’s just we have a new, augmented to build them. And now we have a tweet that’s been drafted for our review, is also a very important thing. We can click “run,” and there are, we’re the manager, we’re able to inspect, we’re able to the work of the AI if we want to. And so this talk, you will be able to access this yourself. And there go. Cool. Thank you, everyone.
(Applause)
So we’ll cut back to slides. Now, the important thing about how we build this, it’s just about building these tools. It’s about teaching the AI how use them. Like, what do we even want it to do when ask these very high-level questions? And to do this, we an old idea. If you go back to Alan Turing’s 1950 paper on the Turing test, he says, you’ll program an answer to this. Instead, you can learn it. You could a machine, like a human child, and then teach it feedback. Have a human teacher who provides rewards and as it tries things out and does things that are either good or bad.
And this is how we train ChatGPT. It’s a two-step process. First, we produce what Turing have called a child machine through an unsupervised learning process. just show it the whole world, the whole internet say, “Predict what comes next in text you’ve never seen before.” And this process imbues it all sorts of wonderful skills. For example, if you’re shown a math problem, the only way to complete that math problem, to say what comes next, that green nine there, is to actually solve the math problem.
But we have to do a second step, too, which is teach the AI what to do with those skills. for this, we provide feedback. We have the AI try multiple things, give us multiple suggestions, and then a human them, says “This one’s better than that one.” And this reinforces not just the specific that the AI said, but very importantly, the whole process that the AI to produce that answer. And this allows it to generalize. allows it to teach, to sort of infer your intent and it in scenarios that it hasn’t seen before, that hasn’t received feedback.
Now, sometimes the things we have teach the AI are not what you’d expect. For example, we first showed GPT-4 to Khan Academy, they said, “Wow, is so great, We’re going to be able to students wonderful things. Only one problem, it doesn’t double-check students’ math. If there’s bad math in there, it will happily pretend that plus one equals three and run with it.” So we had to collect some feedback data. Khan himself was very kind and offered 20 hours of his own time provide feedback to the machine alongside our team. And over the of a couple of months we were able to teach the AI that, “Hey, really should push back on humans in this specific kind of scenario.” And we’ve actually lots and lots of improvements to the models this way. And you push that thumbs down in ChatGPT, that actually is kind like sending up a bat signal to our team say, “Here’s an area of weakness where you should feedback.” And so when you do that, that’s one way that really listen to our users and make sure we’re something that’s more useful for everyone.
Now, providing high-quality feedback a hard thing. If you think about asking a kid to clean their room, if you’re doing is inspecting the floor, you don’t know you’re just teaching them to stuff all the toys the closet. This is a nice DALL-E-generated image, by the way. And the same of reasoning applies to AI. As we move to harder tasks, we will to scale our ability to provide high-quality feedback. But this, the AI itself is happy to help. It’s to help us provide even better feedback and to scale our ability to supervise the machine time goes on. And let me show you what mean.
For example, you can ask GPT-4 a question this, of how much time passed between these two blogs on unsupervised learning and learning from human feedback. And the model two months passed. But is it true? Like, these models are not 100-percent reliable, although they’re getting every time we provide some feedback. But we can actually use the AI to fact-check. And it can check its own work. You can say, fact-check this me.
Now, in this case, I’ve actually given the a new tool. This one is a browsing tool where the model can issue queries and click into web pages. And it actually writes its whole chain of thought as it does it. It says, I’m going to search for this and it actually does the search. It then it finds the publication date the search results. It then is issuing another search query. It’s going click into the blog post. And all of this you do, but it’s a very tedious task. It’s not a thing that humans really want do. It’s much more fun to be in the driver’s seat, to be in this manager’s position where can, if you want, triple-check the work. And out citations so you can actually go and very easily verify piece of this whole chain of reasoning. And it turns out two months was wrong. Two months and one week, was correct.
(Applause)
And we’ll cut back to the side. And thing that’s so interesting to me about this whole process is that it’s this many-step collaboration between a and an AI. Because a human, using this fact-checking is doing it in order to produce data for another AI become more useful to a human. And I think really shows the shape of something that we should to be much more common in the future, where we have humans machines kind of very carefully and delicately designed in how they fit into a problem and we want to solve that problem. We make sure the humans are providing the management, the oversight, the feedback, the machines are operating in a way that’s inspectable trustworthy. And together we’re able to actually create even more trustworthy machines. And I think over time, if we get this process right, we will be able to solve problems.
And to give you a sense of just impossible I’m talking, I think we’re going to be able to rethink almost every aspect of we interact with computers. For example, think about spreadsheets. They’ve been in some form since, we’ll say, 40 years ago with VisiCalc. I don’t think they’ve really changed much in that time. And here is a specific spreadsheet of all the papers on the arXiv for the past 30 years. There’s about 167,000 them. And you can see there the data right here. let me show you the ChatGPT take on how analyze a data set like this.
So we can give ChatGPT to yet another tool, this one a Python interpreter, it’s able to run code, just like a data scientist would. so you can just literally upload a file and ask questions about it. very helpfully, you know, it knows the name of the and it’s like, “Oh, this is CSV,” comma-separated value file, “I’ll parse it for you.” The only here is the name of the file, the column names like saw and then the actual data. And from that it’s able to infer what these columns mean. Like, that semantic information wasn’t in there. It has to sort of, put its world knowledge of knowing that, “Oh yeah, arXiv is site that people submit papers and therefore that’s what these are and that these are integer values and so therefore it’s number of authors in the paper,” like all of that, that’s work for a human to do, the AI is happy to help with it.
Now I don’t even know what want to ask. So fortunately, you can ask the machine, “Can you some exploratory graphs?” And once again, this is a high-level instruction with lots of intent behind it. But I don’t even know what I want. And AI kind of has to infer what I might be interested in. And it comes up with some good ideas, I think. a histogram of the number of authors per paper, time series papers per year, word cloud of the paper titles. All that, I think, will be pretty interesting to see. And the great is, it can actually do it. Here we go, a nice bell curve. see that three is kind of the most common. It’s going to then make nice plot of the papers per year. Something crazy happening in 2023, though. Looks like we were on an and it dropped off the cliff. What could be going there? By the way, all this is Python code, you inspect. And then we’ll see word cloud. So you see all these wonderful things that appear in these titles.
But I’m pretty unhappy about this 2023 thing. It makes this look really bad. Of course, the problem is that the year is not over. So I’m going push back on the machine. [Waitttt that’s not fair!!! 2023 isn’t over. What percentage of papers in 2022 were even posted April 13?] So April 13 was the cut-off date believe. Can you use that to make a fair projection? we’ll see, this is the kind of ambitious one.
(Laughter)
So know, again, I feel like there was more I wanted of the machine here. I really wanted it to notice this thing, maybe it’s little bit of an overreach for it to have of, inferred magically that this is what I wanted. I inject my intent, I provide this additional piece of, know, guidance. And under the hood, the AI is just writing again, so if you want to inspect what it’s doing, it’s very possible. And now, it does the projection.
(Applause)
If you noticed, it even updates the title. I didn’t ask that, but it know what I want.
Now we’ll cut back to the slide again. slide shows a parable of how I think we … A of how we may end up using this technology the future. A person brought his very sick dog to vet, and the veterinarian made a bad call to say, “Let’s just wait see.” And the dog would not be here today had he listened. In the meanwhile, he the blood test, like, the full medical records, to GPT-4, which said, “I not a vet, you need to talk to a professional, here are hypotheses.” He brought that information to a second vet who used it save the dog’s life. Now, these systems, they’re not perfect. cannot overly rely on them. But this story, I think, shows a human with a medical professional and with ChatGPT as a brainstorming partner was able to achieve an that would not have happened otherwise. I think this something we should all reflect on, think about as we consider to integrate these systems into our world.
And one thing believe really deeply, is that getting AI right is to require participation from everyone. And that’s for deciding how we want it to slot in, that’s setting the rules of the road, for what an AI will and won’t do. And if there’s one to take away from this talk, it’s that this just looks different. Just different from anything people had anticipated. And so we all to become literate. And that’s, honestly, one of the reasons released ChatGPT.
Together, I believe that we can achieve the OpenAI mission of that artificial general intelligence benefits all of humanity.
Thank you.
(Applause)
(Applause ends)
Chris Anderson: Greg. Wow. mean … I suspect that within every mind out here there’s feeling of reeling. Like, I suspect that a very large number of people viewing this, look at that and you think, “Oh my goodness, pretty much every single about the way I work, I need to rethink.” Like, there’s new possibilities there. Am I right? Who thinks that they’re to rethink the way that we do things? Yeah, I mean, it’s amazing, but it’s also scary. So let’s talk, Greg, let’s talk.
I mean, I guess my question actually is just how the hell have you this?
(Laughter)
OpenAI has a few hundred employees. Google has thousands employees working on artificial intelligence. Why is it you who’s come up this technology that shocked the world?
Greg Brockman: I mean, the is, we’re all building on shoulders of giants, right, there’s no question. If look at the compute progress, the algorithmic progress, the data progress, all of those really industry-wide. But I think within OpenAI, we made lot of very deliberate choices from the early days. And the first one was just to reality as it lays. And that we just thought really about like: What is it going to take to make progress here? We tried a lot of that didn’t work, so you only see the things that did. And I think that the most thing has been to get teams of people who are different from each other to work together harmoniously.
CA: Can we have the water, the way, just brought here? I think we’re going need it, it’s a dry-mouth topic. But isn’t there also just about the fact that you saw something these language models that meant that if you continue invest in them and grow them, that something at some might emerge?
GB: Yes. And I think that, I mean, honestly, I think the story is pretty illustrative, right? I think that high level, deep learning, like we knew that was what we wanted to be, was a deep lab, and exactly how to do it? I think in the early days, we didn’t know. We tried a lot things, and one person was working on training a model to predict next character in Amazon reviews, and he got a result where — this is a syntactic process, expect, you know, the model will predict where the go, where the nouns and verbs are. But he actually a state-of-the-art sentiment analysis classifier out of it. This could tell you if a review was positive or negative. I mean, today we are like, come on, anyone can do that. But this was the first time that you saw emergence, this sort of semantics that emerged from this underlying syntactic process. And there knew, you’ve got to scale this thing, you’ve got see where it goes.
CA: So I think this helps explain riddle that baffles everyone looking at this, because these things are described as prediction machines. yet, what we’re seeing out of them feels … just feels impossible that that could come from a machine. Just the stuff you showed us just now. And the key idea of emergence that when you get more of a thing, suddenly different things emerge. It all the time, ant colonies, single ants run around, when you bring enough of them together, you these ant colonies that show completely emergent, different behavior. Or a city where a houses together, it’s just houses together. But as you grow number of houses, things emerge, like suburbs and cultural centers and traffic jams. me one moment for you when you saw just something that just blew your mind that you just did not see coming.
GB: Yeah, well, so can try this in ChatGPT, if you add 40-digit numbers —
CA: 40-digit?
GB: 40-digit numbers, the model do it, which means it’s really learned an internal circuit for to do it. And the really interesting thing is actually, if you have it like a 40-digit number plus a 35-digit number, it’ll often get it wrong. And so you can see it’s really learning the process, but it hasn’t fully generalized, right? It’s like can’t memorize the 40-digit addition table, that’s more atoms than there are in universe. So it had to have learned something general, but that it hasn’t really fully yet that, Oh, I can sort of generalize this to adding numbers of arbitrary lengths.
CA: So what’s happened here that you’ve allowed it to scale up and look at an incredible number of pieces text. And it is learning things that you didn’t know it was going to be capable of learning.
GB Well, yeah, and it’s more nuanced, too. So one science we’re starting to really get good at is predicting some these emergent capabilities. And to do that actually, one of the things think is very undersung in this field is sort of engineering quality. Like, we had to rebuild our stack. When you think about building a rocket, every tolerance has to incredibly tiny. Same is true in machine learning. You have get every single piece of the stack engineered properly, and then you can doing these predictions. There are all these incredibly smooth scaling curves. They tell you deeply fundamental about intelligence. If you look at our GPT-4 post, you can see all of these curves in there. And now we’re to be able to predict. So we were able predict, for example, the performance on coding problems. We look at some models that are 10,000 times or 1,000 times smaller. And there’s something about this that is actually smooth scaling, though it’s still early days.
CA: So here is, of the big fears then, that arises from this. If it’s fundamental to what’s happening here, as you scale up, things emerge that you can maybe in some level of confidence, but it’s capable of you. Why isn’t there just a huge risk of something terrible emerging?
GB: Well, I think all of these questions of degree and scale and timing. And I think one people miss, too, is sort of the integration with the world also this incredibly emergent, sort of, very powerful thing too. And so that’s of the reasons that we think it’s so important to deploy incrementally. And so I think what we kind of see right now, if you look at this talk, a lot of I focus on is providing really high-quality feedback. Today, the that we do, you can inspect them, right? It’s very easy to look that math problem and be like, no, no, no, machine, seven was the correct answer. But even summarizing book, like, that’s a hard thing to supervise. Like, how do know if this book summary is any good? You have to read the whole book. one wants to do that.
(Laughter) And so I that the important thing will be that we take this step by step. And we say, OK, as we move on to book summaries, we to supervise this task properly. We have to build up track record with these machines that they’re able to actually carry our intent. And I think we’re going to have to produce even better, efficient, more reliable ways of scaling this, sort of like making the machine aligned with you.
CA: So we’re going to hear later in this session, there are critics say that, you know, there’s no real understanding inside, the system is to always — we’re never going to know that it’s not errors, that it doesn’t have common sense and so forth. it your belief, Greg, that it is true at any one moment, but that the expansion of the and the human feedback that you talked about is going to take it on that journey of actually to things like truth and wisdom and so forth, with a degree of confidence. Can you be sure of that?
GB: Yeah, well, I think that OpenAI, I mean, the short answer is yes, I believe that is where we’re headed. And I think the OpenAI approach here has always been just like, let reality hit you the face, right? It’s like this field is the field of broken promises, all these experts saying X is going to happen, is how it works. People have been saying neural nets aren’t going work for 70 years. They haven’t been right yet. They might be right maybe 70 years one or something like that is what you need. But I think that our approach has been, you’ve got to push to the limits of this technology really see it in action, because that tells you then, oh, here’s how we can move on to a paradigm. And we just haven’t exhausted the fruit here.
CA: I mean, it’s quite a stance you’ve taken, that the right way to do this is to put it out there public and then harness all this, you know, instead just your team giving feedback, the world is now giving feedback. … If, you know, bad things are going to emerge, it is out there. So, you know, the story that I heard on OpenAI when you were founded as nonprofit, well you were there as the great sort of check the big companies doing their unknown, possibly evil thing with AI. And were going to build models that sort of, you know, somehow held accountable and was capable of slowing the field down, if need be. Or least that’s kind of what I heard. And yet, what’s happened, arguably, is the opposite. That your of GPT, especially ChatGPT, sent such shockwaves through the tech world that Google and Meta and so forth are all scrambling catch up. And some of their criticisms have been, you forcing us to put this out here without proper guardrails or die. You know, how do you, like, make the that what you have done is responsible here and not reckless.
GB: Yeah, we think these questions all the time. Like, seriously all the time. And I don’t we’re always going to get it right. But one thing I has been incredibly important, from the very beginning, when were thinking about how to build artificial general intelligence, actually have benefit all of humanity, like, how are you supposed to do that, right? that default plan of being, well, you build in secret, get this super powerful thing, and then you figure out the safety it and then you push “go,” and you hope got it right. I don’t know how to execute that plan. Maybe someone else does. But me, that was always terrifying, it didn’t feel right. And so I think that alternative approach is the only other path that I see, which is that do let reality hit you in the face. And I think do give people time to give input. You do have, these machines are perfect, before they are super powerful, you actually have the ability to see them in action. we’ve seen it from GPT-3, right? GPT-3, we really were afraid that the one thing people were going to do with it was misinformation, try to tip elections. Instead, the number one thing generating Viagra spam.
(Laughter)
CA: So Viagra spam is bad, but there are things that much worse. Here’s a thought experiment for you. Suppose you’re in a room, there’s a box on the table. believe that in that box is something that, there’s a very strong chance it’s something absolutely that’s going to give beautiful gifts to your family and everyone. But there’s actually also a one percent thing in small print there that says: “Pandora.” And there’s a that this actually could unleash unimaginable evils on the world. Do open that box?
GB: Well, so, absolutely not. I you don’t do it that way. And honestly, like, I’ll you a story that I haven’t actually told before, which that shortly after we started OpenAI, I remember I was in Puerto for an AI conference. I’m sitting in the hotel room just looking out over this wonderful water, these people having a good time. And you think about it for a moment, if you choose for basically that Pandora’s box to be five years away or 500 years away, would you pick, right? On the one hand you’re like, well, maybe for you personally, it’s better have it be five years away. But if it gets to be 500 years away and people get time to get it right, which do you pick? And know, I just really felt it in the moment. I like, of course you do the 500 years. My brother in the military at the time and like, he puts his life on the line a much more real way than any of us typing things in computers and developing this technology the time. And so, yeah, I’m really sold on you’ve got to approach this right. But I don’t that’s quite playing the field as it truly lies. Like, if you look the whole history of computing, I really mean it I say that this is an industry-wide or even just like a human-development- of-technology-wide shift. And the more that you of, don’t put together the pieces that are there, right, we’re still making faster computers, we’re still improving algorithms, all of these things, they are happening. And if you don’t them together, you get an overhang, which means that someone does, or the moment that someone does manage to to the circuit, then you suddenly have this very thing, no one’s had any time to adjust, who knows kind of safety precautions you get. And so I think one thing I take away is like, even you think about development of sort of technologies, think about nuclear weapons, people talk about like a zero to one, sort of, change in what humans could do. But actually think that if you look at capability, it’s been quite smooth time. And so the history, I think, of every we’ve developed has been, you’ve got to do it incrementally you’ve got to figure out how to manage it for each moment that you’re it.
CA: So what I’m hearing is that you … the model you us to have is that we have birthed this child that may have superpowers that take humanity to a whole new place. is our collective responsibility to provide the guardrails for this child to collectively teach it to wise and not to tear us all down. Is that basically model?
GB: I think it’s true. And I think it’s important to say this may shift, right? We’ve got to each step as we encounter it. And I think it’s incredibly today that we all do get literate in this technology, figure out how to provide feedback, decide what we want from it. And my is that that will continue to be the best path, but it’s good we’re honestly having this debate because we wouldn’t if it weren’t out there.
CA: Greg Brockman, thank you so much for coming to and blowing our minds.
(Applause)