Sandhini Agarwal: We have now a whole lot of subsequent steps. I undoubtedly assume how viral ChatGPT has gotten has made a whole lot of points that we knew existed actually bubble up and develop into important—issues we need to resolve as quickly as doable. Like, we all know the mannequin continues to be very biased. And sure, ChatGPT is excellent at refusing dangerous requests, however it’s additionally fairly straightforward to write down prompts that make it not refuse what we needed it to refuse.
Liam Fedus: It’s been thrilling to observe the various and artistic purposes from customers, however we’re all the time targeted on areas to enhance upon. We expect that by way of an iterative course of the place we deploy, get suggestions, and refine, we will produce essentially the most aligned and succesful know-how. As our know-how evolves, new points inevitably emerge.
Sandhini Agarwal: Within the weeks after launch, we checked out a number of the most horrible examples that folks had discovered, the worst issues individuals have been seeing within the wild. We type of assessed every of them and talked about how we should always repair it.
Jan Leike: Generally it’s one thing that’s gone viral on Twitter, however we’ve some individuals who really attain out quietly.
Sandhini Agarwal: A whole lot of issues that we discovered have been jailbreaks, which is certainly an issue we have to repair. However as a result of customers should attempt these convoluted strategies to get the mannequin to say one thing dangerous, it isn’t like this was one thing that we utterly missed, or one thing that was very shocking for us. Nonetheless, that’s one thing we’re actively engaged on proper now. Once we discover jailbreaks, we add them to our coaching and testing information. The entire information that we’re seeing feeds right into a future mannequin.
Jan Leike: Each time we’ve a greater mannequin, we need to put it out and take a look at it. We’re very optimistic that some focused adversarial coaching can enhance the state of affairs with jailbreaking lots. It’s not clear whether or not these issues will go away fully, however we predict we will make a whole lot of the jailbreaking much more tough. Once more, it’s not like we didn’t know that jailbreaking was doable earlier than the discharge. I feel it’s very tough to essentially anticipate what the true security issues are going to be with these programs when you’ve deployed them. So we’re placing a whole lot of emphasis on monitoring what persons are utilizing the system for, seeing what occurs, after which reacting to that. This isn’t to say that we shouldn’t proactively mitigate security issues once we do anticipate them. However yeah, it is vitally exhausting to foresee all the pieces that can really occur when a system hits the true world.
In January, Microsoft revealed Bing Chat, a search chatbot that many assume to be a model of OpenAI’s formally unannounced GPT-4. (OpenAI says: “Bing is powered by certainly one of our next-generation fashions that Microsoft custom-made particularly for search. It incorporates developments from ChatGPT and GPT-3.5.”) Using chatbots by tech giants with multibillion-dollar reputations to guard creates new challenges for these tasked with constructing the underlying fashions.
