OG AGI Cyberneticist @daiosai. Ex AI Safety Team lead @googledeepmind. PhD @CambridgeMLG. Will only post about AI-related stuff, never politics. Promise.
London
Joined July 2011
- Tweets2.9K
- Following923
- Followers3.5K
- Likes1.3K
Pinned Tweet
Is it possible to build a decentralized AGI? One made of components owned by different people? Can it be economically inclusive? And can it be made safe?
These are questions I've been asking myself for many years. 1/n
"All science is either physics or stamp collecting"
---Rutherford?
"The philosophers have only interpreted the world, in various ways; the point, however, is to change it."
---Marx
Don't be a stamp collector.
Pedro A. Ortega retweeted
Imagine a world where instead of hiding every model behind closed doors for hacking we instead had things more opened. Perhaps we wouldn't have this week from hell....
Perhaps handing capabilities to a select few while attackers used abliterated models wasn't the best strategy...
this almost feels intentional because now everything is going to be hacked and we are going to have literal weeks of chaos
Pedro A. Ortega retweeted
Hiring PhD student (London, 5 months) at Google DeepMind.
Topic: Steering multi-agent LLM economies via mechanisms for enabling reusable knowledge work.
Looking for strong phd students in Multi agent LLMs, MARL, or mechanism design.
Send cv to [email protected]
Another idea is turning fluid intelligence into crystallized intelligence. If an LLM keeps solving a problem by composing the same actions in the same way, why not turn that composition into an action? Why keep rediscovering the same solution? 17/n