The current bottleneck is political will, not research
Alignment Forum · Charbel-Raphaël · 2026-07-11
Charbel-Raphaël, an AI governance think-tank director, argues that AI safety's main bottleneck is not a shortage of research or policy ideas but a lack of political will among policymakers, and calls for rebalancing the field's investment heavily toward advocacy and direct engagement.
Appears in
Extraction
Topics: ai-governanceai-safety-advocacypolitical-willai-policyexistential-risk-communication
Claims
- Most of the top 100-1,000 most influential global policymakers have never had a serious conversation about catastrophic AI risk, and this absence — not a shortage of research — is the main reason AI safety measures are not being implemented.
- The US AI safety field has approximately 3.6 researchers per advocate, a ratio the author argues should be inverted to closer to 3 advocates per researcher.
- Out of 1,534 written submissions to the UN Global Dialogue on AI, only 1 mentions 'takeover' and fewer than 1% mention existential risks, even though these are the risks the submitting organizations privately prioritize.
- AI companies secured 7 times as many meetings with the European Commission on AI as civil society in 2023, and the entire AI safety governance field is roughly two orders of magnitude smaller than the climate advocacy ecosystem.
- Ryan Greenblatt's estimates suggest that strong political will (Plan A versus Plan D) could reduce conditional AI takeover risk from approximately 45% to 7%, meaning almost all risk reduction is purchased by political will rather than technical research.
Key quotes
'We already know enough to act. I wish we were in a world where research was the bottleneck, but the main constraint on AI safety is no longer a shortage of clever policy ideas: best practices already exist and are not being applied or enforced.'
Buck Shlegeris: 'Five years ago I thought of misalignment risk from AIs as a really hard problem that you'd need some really galaxy-brained fundamental insights to resolve. Whereas now, to me the situation feels a lot more like we just really know a list of 40 things where, if you did them — none of which seem that hard — you'd probably be able to not have very much of your problem. But I've just also updated drastically downward on how many things AI companies have the time/appetite to do.'
'In research, novelty is the main value. In governance it's almost the opposite: several people pushing the same thing, independently, is what works. There is no h-index for minds changed.'