-
other
2026-04-24 ~
↺ غيّر موقفه
After being pushed out of Aleph Alpha (ahead of Cohere's takeover) sums up his exit: 'I'm out.'
«I'm out.»
-
other
2025-10-02
Against wholesale hyperscaler adoption: it turns you into a user of technology built and governed elsewhere.
«The hyperscaler model... turns you into a sophisticated user of technology built and governed elsewhere.»
-
products
2025-02-23
Won't become a cloud provider — wants customers free without lock-in; knowledge in documents, not LLM parameters.
«We will never become a cloud provider. I want my customers free and not locked in.»
-
open source
2025-02-23
↺ غيّر موقفه
Confirms the pivot: no need to build another Llama or DeepSeek — they already exist.
«We don't have to build another Llama or DeepSeek because they're already out there.»
-
products
2025-02-23
Positions Aleph Alpha as the 'operating system'/foundation for enterprises and governments to build sovereign AI.
«be this operating system for enterprises and governments to build their own sovereign AI strategy.»
-
compute
2024-10-03
The LLM business is hard to make economically viable; it's only easy to wire lots of money to Nvidia.
«It's hard to make it make sense economically. It's not hard to wire a lot of money to Nvidia.»
-
products
2024-09-05
↺ غيّر موقفه
Key pivot: a European LLM alone doesn't justify the investment — from the foundation-model race to an applied platform (PhariaAI).
«Just having a European LLM is not sufficient as a business model. It doesn't justify the investment.»
-
regulation
2024-01-30
The agreed EU AI Act is 'not horrible' (earlier drafts were worse), but compliance costs will bite industry.
«not horrible»
-
safety
2024-01-24
AI gives humans superpowers, but also empowers criminals and those who want to influence elections.
«AI gives humans superpowers, but also to people who want to influence elections, bring people down, do crime.»
