Jeff Dean: belief

Transcript context

…Woo. Thank God Google's doing well as well. So these days we do publish a little less of what we're doing. There's always this trade-off: should we publish exactly what we're doing right away? Should we put it in the next stages of research and then roll it out into production Gemini models and not publish it at all? Or is there some intermediate point? And for example, in our computational photography work in Pixel cameras, we've often taken the decision to develop interesting new techniques, like the ability to do super good night sight vision for low-light situations or whatever, put that into the product and then published a real research paper about the system that does that after the product is released. Different techniques and developments have different treatments. Some things we think are super critical we might not publish. Some things we think are really interesting but important for improving our products; we'll get them out into our products and then make a decision: did we publish this or do we give kind of a lightweight discussion of it, but maybe not every last detail? Other things I think we publish openly and try to advance the field and the community because that's how we all benefit from participating. I think it's great to go to conferences like NeurIPS last week with 15,000 people all sharing lots and lots of great ideas. We publish a lot of papers there as we have in the past, and see the field advance is super exciting. How would you account for... so obviously Google had all these insights internally rather early on, including the top researchers. And now Gemini 2 is out. We didn't get a chance much to talk about it, but people know it's a really great model.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.