Executive summary: The author reviews the AI safety landscape and argues that neglected areas—especially AI existential-risk (x-risk) policy advocacy and ensuring transformative AI (TAI) goes well for animals—deserve more attention, highlighting four priority projects: engaging policymakers, drafting legislation, making AI training more animal-friendly, and developing short-timeline plans for animal welfare.
Key points:
Technical safety research is comparatively well-funded, while AI x-risk advocacy is neglected; Dickens prioritizes advocacy over research, despite risks of backfire or slowing progress.
Short timelines (25–75% chance of TAI within 5 years) make quick-payoff advocacy more urgent than long-horizon research.
Top recommended projects: (a) talk to policymakers about AI x-risk, (b) draft AI safety legislation, (c) advocate for LLM training that includes animal welfare, and (d) design/evaluate short-timeline animal welfare interventions.
Post-TAI animal welfare may be less critical than human survival but remains cost-effective and underfunded relative to its importance.
Non-alignment issues (digital minds, S-risks, moral error, gradual disempowerment) are highly important but judged intractable under short timelines, so not prioritized here.
General recommendations: advocacy should explicitly emphasize extinction and misalignment risks, prioritize work useful under short timelines, and consider slowing AI development as a cross-cutting solution.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.
Executive summary: The author reviews the AI safety landscape and argues that neglected areas—especially AI existential-risk (x-risk) policy advocacy and ensuring transformative AI (TAI) goes well for animals—deserve more attention, highlighting four priority projects: engaging policymakers, drafting legislation, making AI training more animal-friendly, and developing short-timeline plans for animal welfare.
Key points:
Technical safety research is comparatively well-funded, while AI x-risk advocacy is neglected; Dickens prioritizes advocacy over research, despite risks of backfire or slowing progress.
Short timelines (25–75% chance of TAI within 5 years) make quick-payoff advocacy more urgent than long-horizon research.
Top recommended projects: (a) talk to policymakers about AI x-risk, (b) draft AI safety legislation, (c) advocate for LLM training that includes animal welfare, and (d) design/evaluate short-timeline animal welfare interventions.
Post-TAI animal welfare may be less critical than human survival but remains cost-effective and underfunded relative to its importance.
Non-alignment issues (digital minds, S-risks, moral error, gradual disempowerment) are highly important but judged intractable under short timelines, so not prioritized here.
General recommendations: advocacy should explicitly emphasize extinction and misalignment risks, prioritize work useful under short timelines, and consider slowing AI development as a cross-cutting solution.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.