Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

blog.google

Post reply on HN
Print view View original

621–630 of 989 posts

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#621

Post by ActionHank »

squidbeak wrote: post #284
GodelNumbering wrote: post #199

So, in last several months, all the prominent names Google lost: Demis Hassabis (technically still with google but these things are usually presented with a spin), Jeff Dean, Sanjay Ghemawat, Oriol Vinyals, Quoc Le, Noam Shazeer, John Jumper, Jonas Adler, Alexander Pritzel, David Silver, Denny Zhou, Fernando Pereira, Alex Turner And all the prominent names Google gained: NULL Combined with no gemini frontier GA relea…

Gemini 3 Pro was legitimately frontier and released 8 1/2 months ago, not 14.

[flagged]

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#622

Post by grogenaut »

underlipton wrote: post #517
Nifty3929 wrote: post #441

Earlier quoted context omitted.

Dynamite has been incredibly beneficial for humanity.

Indirectly. Direct application of dynamite to the human body is considerably more fraught an event than direct application of penicillin. As the fundamental goal of technology is to extend the capability of the human body, it's natural to implicitly and primarily consider what that extended capability can do to another human body. It's nice that we have tunnels through mountains and bedrock.

Dynamite is ~20-60% nitro glycerine and the rest "dope" aka a stabilizer which could be sawdust for instance [https://en.wikipedia.org/wiki/Dynamite].

If you have heart issues, you're likely taking nitro daily for chest pain.

So saying it's bad is 40% wrong at least. Makes ya think.

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#623

Post by lesuorac »

kridsdale1 wrote: post #481
frollogaston wrote: post #434

Earlier quoted context omitted.

Only the GFG models know anything internal. Regular Gemini isn't trained on any of that. And GFG is a much older base model, so people use the regular one. If the tools seem to handle google3 code ok, it's only because of skills and not the model itself, and then you run into issues with skill bloat. Sometimes the A/B test would give me the bad model of the day that'd try to grep all of piper. Start in a blank direct…

Your information is indeed out of date.

Dunno about the Boq stuff but it regularly tries to run `git status` in a fig workspace ...

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#624

Post by swrrt »

fooker wrote: post #457
frollogaston wrote: post #294

Earlier quoted context omitted.

Part of me thinks Google's entire problem is crappy internal tooling, not really an anti-innovation environment. Just making a dev take 2x as long to get something done has a bigger effect than you'd think. With LLMs it's more like 10x now because even Gemini doesn't understand Google-internal tooling.

Google has close to the best internal tooling in the industry for a decade or so. Then the Google engineers who joined Facebook missed it so much that they built a better replacement.

Sounds like a big-company issue, large scale lead to large burden and complexity.

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#626

Post by RealWed5 »

firasd wrote: post #320
thorncorona wrote: post #315

Earlier quoted context omitted.

Gemini isn't as heavily aligned as OAI / Ant models ..?

If you just talk to it over API (no web search) the Gemini models are extremely resistant to thinking the user may be living in a universe outside their training data. Try to discuss any news etc and they assume it’s fake or fiction

[dead]

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#627

Post by bushbaba »

Cidan wrote: post #550
frollogaston wrote: post #294

Earlier quoted context omitted.

Part of me thinks Google's entire problem is crappy internal tooling, not really an anti-innovation environment. Just making a dev take 2x as long to get something done has a bigger effect than you'd think. With LLMs it's more like 10x now because even Gemini doesn't understand Google-internal tooling.

Google's internal tooling is still, hands down, better than anything that exists for the scale they operate at, and it's not even close. Google's processes, however, is hands down the worst thing to exist for the scale they operate at, and it's not even close.

the processes are a result of repeated lawsuit

Re: Changes at Google DeepMind: Demis Hassabis from CEO to Chair, Jeff Dean departs

#630

Post by sinuhe69 »

My read is that AGI is a long shoot and nobody expects it in the next 10 years. They will push for it as always. But the cost of AI is exploding and requires a strong hand and a pragmatic approach if you don’t want to go down with it once the bubble bursts.

Thus, efficiency and pragmatic enterprise application is the next focus. And for that, the old guards are not necessarily the best. Google needs new blood for this venture, preferably from China with their practical engineering view. Every big AI lab is on the verge. Steering the right way will help avoid a catastrophe and come out stronger than ever. A weak captain could sand the ship and drag down who knows what with them.

Post reply on HN