p-zero research
We measure how fast AIs are recursively self-improving and share that data with the world. Towards this, we develop open-ended research and engineering tasks, and measure how well AIs can carry them out. This data is necessary for understanding current capabilities, developing accurate forecasts, and making well-calibrated decisions in governance, deployment, and safety research. Together, these decisions determine how well misalignment and concentration-of-power risks are managed. We're hiring exceptional engineers and researchers. Contact us at hello@pzeroresearch.com
p-zero research p-zero research We measure how fast AIs are recursively self-improving and share that data with the world. Towards this, we develop open-ended research and engineering tasks, and measure how well AIs can carry them out. This data is necessary for understanding current capabilities, developing accurate forecasts, and making well-calibrated decisions in governance, deployment, and safety research. Together, these decisions determine how well misalignment and concentration-of-power risks are managed. We're hiring exceptional engineers and researchers. Contact us at [email pro
saved by
related reading
- When AI builds itself \ Anthropicanthropic.com
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- Automated Weak-to-Strong Researcheralignment.anthropic.com
- Focus areas for The Anthropic Instituteanthropic.com
- Physical Intelligence (π)pi.website
- Physical Intelligence (π)physicalintelligence.company
- How Should the US Prepare for Increasingly Automated AI R&D? | IFPifp.org
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Toward an O*NET for AI R&Depochai.substack.com
- Will AI R&D Automation Cause a Software Intelligence Explosion?forethought.org
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- METRmetr.org