# Findings: more References

Canonical: https://machines-on-machines.anth.us/reference-lists/findings/untagged/c416ab6b09c/

- [LLM-based NPCs in Video Games: A Systematic Mapping Study](https://arxiv.org/abs/2611.05427)
- [PerturbAI and OpenAI Unveil First-of-Its-Kind AI Reasoning Agents to Scale Biomedical Discovery](https://www.prnewswire.com/news-releases/perturbai-and-openai-unveil-first-of-its-kind-ai-reasoning-agents-to-scale-biomedical-discovery-302914929.html)
- [Using thousands of AI agents to uncover new leads in biological data](https://openai.com/index/scaling-ai-research-for-complex-biological-data/)
- [Are Large Pre-Trained Language Models Leaking Your Personal Information?](https://aclanthology.org/2022.findings-emnlp.148.pdf)
- [MCP connectors are not a context engine](https://getunblocked.com/blog/mcp-connectors-are-not-a-context-engine/?utm_source=bytebytego&utm_medium=email&utm_campaign=primary_20261104)
- [InfoQ eMag 123: Securing the AI Stack, From Model to Production](https://ress.infoq.com/minibooks/secure-ai-stack-model-production/en/pdf/InfoQ-eMag-123-Securing-the-AI-Stack-From-Model-to-Production-1774423828257.pdf)
- [Vercel Security Review](https://vercel.com/blog/vercel-security-review)
- [Announcing AWS Well-Architected Agent (preview)](https://aws.amazon.com/blogs/aws/announcing-aws-well-architected-agent-an-ai-powered-intelligence-to-optimize-your-cloud-environment-preview/)
- [On the (Im)possibility of Obfuscating Programs - Springer Nature](https://link.springer.com/chapter/10.1007/3-540-44647-8_1)
- [MLReplicate: Benchmarking Autonomous Research Systems for ...](https://arxiv.org/pdf/2605.16616)
- [When prompts become shells: RCE vulnerabilities in AI ...](https://www.microsoft.com/en-us/security/blog/2026/05/07/prompts-become-shells-rce-vulnerabilities-ai-agent-frameworks/)
- [Evaluating foundation model integration strategies for detecting PII in java software engineering pipelines - Empirical Software Engineering](https://link.springer.com/article/10.1007/s10664-026-10919-y)
- [Guardrails and moderation — Study · AddressHate](https://www.addresshate.org/research/studies/guardrail-moderation)
- [Stealing Reasoning Traces from Proprietary LLM APIs](https://research.snyk.io/blog/stealing-reasoning-traces/)
- [Red-Teaming Coding Agents from a Tool-Invocation Perspective: An Empirical Security Assessment](https://researchportal.hkust.edu.hk/en/publications/red-teaming-coding-agents-from-a-tool-invocation-perspective-an-e/)
- [Crowdsourced audit of Twitter’s recommender systems - Scientific Reports](https://www.nature.com/articles/s41598-023-43980-4)
- [Three AI Coding Agents, One GitHub Issue: CI/CD Secrets Exposed](https://labs.cloudsecurityalliance.org/research/csa-research-note-ai-coding-agent-cicd-secrets-20260808-csa/)
- [MCP Auto-Execution: AI Coding Assistants and Credential Theft](https://labs.cloudsecurityalliance.org/research/csa-research-note-mcp-ai-coding-assistant-credential-theft-2/)
- [The Authority Benchmark: do AI agents follow the rules they are given? | Plaw](https://plaw.io/research)
- [LFM3 Harness Deep Dive: How a Sub-2B ...](https://local-ai-zone.github.io/blog/lfm3-harness-deep-dive.html)
- [Frontiers | Poisoning-resistant signature sharing for hybrid intrusion detection using a permissioned blockchain ledger](https://www.frontiersin.org/articles/10.3389/fbloc.2026.1960105/full)
- [TraceTrial-Bench: How Reliable Are Agent Harnesses Across 3,420 Runs?](https://mer.vin/news/tracetrial-bench-agent-harness-reliability/)

[Previous](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c609892035c/index.md)
[Next](https://machines-on-machines.anth.us/reference-lists/findings/untagged/c25a519ffe5/index.md)