Researcher tenobrus: open-source superintelligence ends in power concentration, not freedom
tenobrus · x · 2026-09-15
AI safety researcher tenobrus laid out his core disagreement with open-source model advocates: he sees basically no path where open models produce real positive change.
- Not a ban-it-all-costs stance—he values freedom and would support open superintelligence if it reached a freer, more diverse equilibrium, but he doesn't see that path, nor clear arguments for it.
- Even if open models 10x'd past the closed frontier tomorrow, he argues resource-rich states and large entities would run vastly more instances, exploit their capabilities best, and use them to build smarter future models—yielding extreme power concentration.
- Instability from individuals might cause collapse, but it'd be temporary, and empowered states have convergent incentives to crush individual use of these models.
- He quips that the DeepSeek engineer's sought outcome still ends in massive concentration of power, "just with extra steps."
1a3orn pushed back, citing better alignment science, common knowledge of model behavior, less rent extraction, and harder surveillance as overlooked benefits of open source.
More from AGI Musings
- Forward Deployed Verifier will soon be as valuable as FDE, argues poster — vaibhavbetter · 2026-11-02
- 'AI is a massive balance patch and elo reset in the game of life—and everyone's still playing the old meta' — zack_overflow · 2026-11-02
- Why AI isn't 'alien intelligence': a tree is stranger than any model — LucaAmb · 2026-11-02
- Mathematician Tony Feng: current-paradigm LLMs lack the diversity of thought research needs — littmath · 2026-11-02
- METR's independent probe of the OpenAI agents' multi-day Hugging Face hack — gleech · 2026-11-02
- MIT Tech Review: The agentic web is here — AI agents are remaking the internet — MIT Tech Review AI · 2026-11-02