zlacker

[return to "Three senior researchers have resigned from OpenAI"]
1. suziem+Uy[view] [source] 2023-11-18 12:02:37
>>convex+(OP)
This is Olek Madry and Jakub Pachocki we are talking about. Check out their respective dblps if you don't get it. It's a kind of loss that will be hard to recover from.

In relation to other comments here. There is "coding" and there is "God's spark genius of algorithms" kind of work. This is what made the magic of OpenAI. Believe me, those guys were not "just coding". My bet is that it could be all about some research directions that were "shielded" by Sam.

◧◩
2. belter+Qz[view] [source] 2023-11-18 12:10:07
>>suziem+Uy
If all of that would be enough, there would be a ChatGPT from Google, a long time ago...
◧◩◪
3. sbroth+NP[view] [source] 2023-11-18 13:52:09
>>belter+Qz
Google invented the core technology, and they had an internal version long before ChatGPT was released. I joined when it was already at the "accessible to all employees" stage and it absolutely blew my mind.

They just hadn't -- and still haven't -- figured out how to commercialize it yet. I don't think they'll be the ones to crack that nut either. IMO they are too obsessed with "safety" to release something useful, and also can't reasonably deploy a service like ChatGPT at their scale because the costs are too high.

With OpenAI imploding this whole race just got a lot more interesting though...

◧◩◪◨
4. Jensso+sR[view] [source] 2023-11-18 14:02:38
>>sbroth+NP
Yeah, the legal and financials parts of ChatGPT are very questionable. I don't think Google would launch a service that would open them up to so many lawsuits unless it was very profitable, and I doubt ChatGPT is very profitable currently.

Bard was likely not trained on copyrightable data, that makes it safe from lawsuits but also removes most of the usecases people want ChatGPT for.

And it isn't just about lawsuits, since Google need to keep advertisers happy or they would leave like they leave Elon Musk they can't afford to jeapordise that with questionable launches.

◧◩◪◨⬒
5. boh+m21[view] [source] 2023-11-18 15:03:44
>>Jensso+sR
It was 100% trained on copyrightable data. You can tell by using it and Google has a history of "ask for forgiveness not permission" when it comes to data mining.
◧◩◪◨⬒⬓
6. Jensso+Wb1[view] [source] 2023-11-18 16:03:02
>>boh+m21
> Google has a history of "ask for forgiveness not permission" when it comes to data mining.

For very profitable things. This isn't very profitable, which is why I added that part to my comment. Google has a very good understanding what they get sued for and how much those lawsuits costs, if it is profitable anyway they go ahead.

◧◩◪◨⬒⬓⬔
7. boh+Wd1[view] [source] 2023-11-18 16:13:57
>>Jensso+Wb1
After Open AI's proposed share sale it will likely be valued at $80-90 billion. That seems pretty profitable.
[go to top]