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