# Findings: more References

Canonical: https://machines-on-machines.anth.us/reference-lists/findings/untagged/c389b4e0093/

- [Safety of a large language model-based clinical decision support system in African primary healthcare](https://www.nature.com/articles/s44360-026-00082-5)
- [Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models](https://arxiv.org/abs/2609.12586)
- [Ironies of Automation](https://tc.ifac-control.org/4/1/newsletter/ironies-of-automation/view)
- [An Investigation of the Therac-25 Accidents](https://web.mit.edu/6.033/2004/wwwdocs/papers/Therac_1.html)
- [Moral Crumple Zones: Cautionary Tales in Human-Robot Interaction](https://doi.org/10.17351/ests2019.260)
- [Autoresearch: AI agents running research on single-GPU nanochat training automatically](https://github.com/karpathy/autoresearch)
- [Collision Between Vehicle Controlled by Developmental Automated Driving System and Pedestrian](https://www.ntsb.gov/investigations/Pages/HWY18MH010.aspx)
- [Valve's October Next Fest most-played demos](https://store.steampowered.com/sale/nextfestmostplayed/?l=english)
- [Hyperagents](https://arxiv.org/abs/2603.19461)
- [Human error: models and management](https://www.bmj.com/content/320/7237/768)
- [AutoResearch at Production Scale: Failure Modes and a Multi-Agent Framework](https://arxiv.org/abs/2609.30541)
- [Constitutional AI: Harmlessness from AI Feedback](https://arxiv.org/abs/2212.08073)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://arxiv.org/abs/2305.16291)
- [Venn Studios compares October Next Fest disclosures](https://venn-studios.com/research/next-fest-ai-disclosure/)
- [Automated Researchers Can Mitigate Well-Characterized Alignment Failures](https://alignment.anthropic.com/2026/automated-alignment-researchers/)

[Previous](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c696eec3f3c/index.md)
[Next](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c58d51e39e1/index.md)