News for hackers. Stuff that matters.
Re:DeepMind AI is as fast as humans at solving previously un... by pjkundert on Saturday February 11, 2023 @11:11AM (#34753575)
The great thing about the current-gen LLMs — anybody will be able to run one on the next gen of hardware. We’re within one order of magnitude of the capacity right now on retail hardware.

The moment of transition from “data-centric” (big centralized system/database) to “agent-centric” (locally stored/run systems and identity, sharing arbitrary data/storage) has arrived.

GDPR, LLM “guard rails”, … — only the plebs will be affected by those.

Re:DeepMind AI is as fast as humans at solving previously un... by TigeriusKirk on Saturday February 11, 2023 @11:56AM (#34754008)
I believe you could run GPT-2 at home now.

Remember when OpenAI claimed that was too much power to allow open access to?

Re:DeepMind AI is as fast as humans at solving previously un... by pjkundert on Saturday February 11, 2023 @01:12PM (#34754687)
Benevolent overlords patting the filthy masses on the head.
Re:DeepMind AI is as fast as humans at solving previously un... by nateburke on Saturday February 11, 2023 @12:29PM (#34754311)
> The moment of transition from “data-centric” (big centralized system/database) to “agent-centric” (locally stored/run systems and identity, sharing arbitrary data/storage) has arrived.

I agree. Use cases such as LLMs-as-backend-of-my-web-stack will move a lot of compute back to the edge of the network, eventually.