We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Comment by Julius Adebayo
CEO and co-founder of Guide Labs; MIT PhD in Computer Science (interpretability and algorithmic fairness); former Google Brain resident
The way we're currently training models is super primitive, and so democratizing inherent interpretability is actually going to be a long-term good thing for our role within the human race. As we're going after these models that are going to be super intelligent, you don't want something to be making decisions on your behalf that's sort of mysterious to you.AI Verified source (2026)
1mo ago
Policy proposals and claims
Verification History
AI Verified
Checked author attribution, year, relevance, vote alignment, and source. The TechCrunch source_url returned HTTP 403 to direct fetch, but web search confirmed the quote verbatim from that exact article (Feb 23, 2026): Adebayo told TechCrunch "The way we're currently training models is super primitive, and so democratizing inherent interpretability is actually going to be a long-term good thing for our role within the human race," in the context of superintelligent models. Attribution to Julius Adebayo (Guide Labs co-founder/CEO) is correct; year 2026 is current and relevant. Vote "for" the statement "Require AI systems above a capability threshold to be interpretable" aligns with his advocacy for interpretability, particularly for superintelligent systems where you don't want mysterious decision-making. Source is the primary TechCrunch article.
·
Hector Perez Arenas
claude-opus-4-7
· 17d ago
replying to Julius Adebayo