Quoting Jeremy Howard
Jeremy Howard proposes a governance model where the leading AI lab voluntarily restricts its own frontier research access to top models while distributing them to competitors, preventing capability concentration and power imbalance. He contrasts this with Anthropic's current stance of retaining exclusive research access to its best models while blocking others. The proposal surfaces a core tension in AI safety: whether competitive parity or centralized control better mitigates risks from recursive self-improvement.
Modelwire context
Analyst takeHoward's framing inverts the usual safety argument: rather than treating restricted access as a safety feature, he treats it as the risk itself, specifically the risk that one organization accumulates disproportionate leverage over recursive self-improvement cycles before any external check exists.
This is largely disconnected from recent activity in our archive, as we have no prior coverage to anchor it to. It belongs, however, to a broader ongoing debate about whether safety and openness are complementary or opposed, a tension that has surfaced repeatedly around Anthropic's public benefit corporation structure and Meta's open-weight releases. Howard's proposal sits at the intersection of those two camps, arguing that voluntary redistribution by the frontier leader is more stabilizing than either full openness or full concentration. That framing puts pressure on Anthropic specifically, since its safety-first branding is the thing Howard is contesting.
Watch whether Anthropic responds publicly to Howard's framing in the next 60 days, either through a policy document or a direct rebuttal from Dario or Daniela Amodei. A non-response is itself informative, since the proposal is pointed enough that silence functions as a position.
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsJeremy Howard · Anthropic · Simon Willison
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. The full content lives on simonwillison.net. If you’re a publisher and want a different summarization policy for your work, see our takedown page.