And so I, I imagined that we could do something similar with potentially additional levels of access for the AIEC team. We talked about this earlier. I would want to talk to the team that made it as a researcher access program and see what other kinds of offerings might ...

Okay, Emma, nice to meet you. Thanks for being our partner. Yeah, so I think with that clarification that you provided to Anita I know that we actually already have a researcher access program. So we’ve worked with external researchers to red team to our products to study areas related ...

The searching and using of links is a general purpose approach. We have classifiers that allow our models to determine if for example data that it needs or the question being asked references stuff that would need the latest and greatest information. But for things like pointing people to canonical ...

Yes, we have a team called Global Affairs led by Anna Makanju. And there’s a specific elections focused team there. And right now, we’re aware that there’s more elections happening in 2024 that have ever happened on record. So they’re constantly working with civil society organizations and local governments. A ...

We have seen promising initial results but that’s kind of a work in progress and open research area that we’re hoping to share more about soon. Also, outside of the image domain, ChatGPT is increasingly integrating with existing source of information. For example, users can access real time news reporting ...

On the last piece for content produced by general AI tools. This is also something so I think Anita already mentioned this, I think Max is also doing this. So for our DALL·E image generation technology we’ve also added the C2PA protocol, that coalition for content provenance and authenticity, which ...

I think that’s something that will really require ongoing dialogue. I know that a lot of that’s already happening between our companies. So excited to see that continue.

For us, a lot of the bad behavior happens when people are trying to use the models to create GPTs to do those things that I just outlined, which we don’t allow. But also there’s of course people who are using our API, or just ChatGPT to generate pieces of ...

We don’t allow builders to create chatbots that pretend to be real people, such as political candidates, or celebrities, or institutions like governments. We don’t allow applications that deter people from participating in democratic processes. And we have teams that focus on also enforcing these usage policies to monitor what’s ...

Yes. So, Audrey and the Moda team, thank you for running these alignment assemblies. Really interesting and important topic. From the top - so for interaction in generative AI, clearly marketing them as such while setting and labeling the sources. I think I’ll answer this one and the second one ...

Sure. Hi, everyone. My name is Teddy Lee. I’m a member of the OpenAI team. Great to be here.