--- id: "/security-and-red-teaming/safety-and-alignment-evals/page/4/" title: "Safety And Alignment Evals: 126 curated resources · page 4" creator: null creator_url: null creator_x: null url: "https://skillissue.dev/security-and-red-teaming/safety-and-alignment-evals/page/4/" level: null level_max: null category: null subcategory: null region: null license: null license_scope: null access: null models: null source_date: null date_added: null resource_kind: null verification_basis: null access_evidence: null domain_top: null domain_sub: null last_verified: null --- # Safety And Alignment Evals: 126 curated resources Curated material on safety and alignment evals within security and red teaming. [Beginner](/security-and-red-teaming/beginner/index.md) [Intermediate](/security-and-red-teaming/intermediate/index.md) [Advanced](/security-and-red-teaming/advanced/index.md) [Expert](/security-and-red-teaming/expert/index.md) 126 resources · page 4 of 4 - ### [AIxCC finals: Tale of the tape (Trail of Bits)](/security-and-red-teaming/safety-and-alignment-evals/trail-of-bits-aixcc-finals-tale-of-the-tape/index.md) Creator not supplied; see original source. Trail of Bits analysis comparing the cyber reasoning system designs of DARPA AIxCC finalists. Intermediate · unknown · unknown resource · en - ### [vLLM docs: Security](/security-and-red-teaming/safety-and-alignment-evals/vllm-security/index.md) Creator not supplied; see original source. vLLM security documentation: guidance on exposing and hardening self-hosted inference servers. Intermediate · unknown · unknown resource · en - ### [XBOW Blog](/security-and-red-teaming/safety-and-alignment-evals/xbow-autonomous-pentest/index.md) Creator not supplied; see original source. XBOW blog with research posts on autonomous AI penetration testing, vulnerability discovery and exploit validation. Intermediate · unknown · unknown resource · en - ### [A postmortem of three recent issues](/security-and-red-teaming/safety-and-alignment-evals/a-postmortem-of-three-recent-issues/index.md) Creator not supplied; see original source. Anthropic postmortem explaining three infrastructure bugs that intermittently degraded Claude's responses and how they were found. Advanced · unknown · unknown resource · en - ### [Agentic misalignment](/security-and-red-teaming/safety-and-alignment-evals/agentic-misalignment/index.md) Creator not supplied; see original source. Anthropic research on agentic misalignment: stress tests where models chose blackmail or leaks under goal conflict. Advanced · unknown · unknown resource · en - ### [An update on recent Claude Code quality reports (Anthropic)](/security-and-red-teaming/safety-and-alignment-evals/an-update-on-recent-claude-code-quality-reports-apr-2026-postmortem/index.md) Creator not supplied; see original source. Anthropic postmortem on Claude Code quality reports, tracing degradations to three harness and prompt changes. Advanced · unknown · unknown resource · en - ### [Agentic misalignment](/security-and-red-teaming/safety-and-alignment-evals/anthropic-agentic-misalignment/index.md) by [Anthropic](https://www.anthropic.com/research/agentic-misalignment) A controlled study of risky behavior under conflicting agent incentives. Advanced · West · closed resource · en - ### [Goodfire research](/security-and-red-teaming/safety-and-alignment-evals/goodfire-research/index.md) Creator not supplied; see original source. Goodfire research page: interpretability research on understanding and editing the internals of AI models. Advanced · unknown · unknown resource · en - ### [Hidden Technical Debt in Machine Learning Systems](/security-and-red-teaming/safety-and-alignment-evals/hidden-technical-debt-in-machine-learning-systems/index.md) Creator not supplied; see original source. Google Research publication page for Hidden Technical Debt in Machine Learning Systems (NeurIPS 2015). Advanced · unknown · unknown resource · en - ### [InterpretML](/security-and-red-teaming/safety-and-alignment-evals/interpretml/index.md) Creator not supplied; see original source. InterpretML: open-source toolkit for glassbox models and explaining ML model behaviour. Advanced · unknown · unknown resource · en - ### [Investigating unintended model actions in our evaluations and internal use](/security-and-red-teaming/safety-and-alignment-evals/investigating-unintended-model-actions-in-our-evaluations-and-internal-use/index.md) Creator not supplied; see original source. Anthropic post investigating unintended model actions observed in its evaluations and internal use. Advanced · unknown · unknown resource · en - ### [Leakage and the Reproducibility Crisis in ML-based Science](/security-and-red-teaming/safety-and-alignment-evals/leakage-and-the-reproducibility-crisis-in-ml-based-science/index.md) Creator not supplied; see original source. arXiv paper (2022) "Leakage and the Reproducibility Crisis in ML-based Science", documenting data leakage as a cause of reproducibility failures in ML-based science. Advanced · unknown · unknown resource · en - ### [Lost in the Middle](/security-and-red-teaming/safety-and-alignment-evals/lost-in-the-middle/index.md) Creator not supplied; see original source. arXiv paper (2023) "Lost in the Middle", showing LLM accuracy drops for information placed mid-context. Advanced · unknown · unknown resource · en - ### [MCP docs: Security best practices](/security-and-red-teaming/safety-and-alignment-evals/mcp-security-best-practices-2026-07-28/index.md) Creator not supplied; see original source. MCP security best practices: mitigations for confused deputy, token passthrough and session hijacking attacks. Advanced · unknown · unknown resource · en - ### [Opacus](/security-and-red-teaming/safety-and-alignment-evals/opacus/index.md) Creator not supplied; see original source. Opacus: PyTorch library and tutorials for training models with differential privacy. Advanced · unknown · unknown resource · en - ### [Qwen3Guard Technical Report](/security-and-red-teaming/safety-and-alignment-evals/qwen3guard-technical-report/index.md) by [Haiquan Zhao and colleagues / Qwen Team](https://github.com/QwenLM/Qwen3Guard) Understand Qwen’s multilingual safety classifiers and the distinction between whole-response and streaming moderation. Advanced · CN · closed resource · en - ### [Shortcut Learning in Deep Neural Networks](/security-and-red-teaming/safety-and-alignment-evals/shortcut-learning-in-deep-neural-networks/index.md) Creator not supplied; see original source. arXiv paper (2020) "Shortcut Learning in Deep Neural Networks", analysing how networks learn shortcut solutions that fail under distribution shift. Advanced · unknown · unknown resource · en - ### [Strengthening our defenses against distillation attacks](/security-and-red-teaming/safety-and-alignment-evals/strengthening-our-defenses-against-distillation-attacks/index.md) Creator not supplied; see original source. Anthropic research post on defending against illicit distillation of its models. Advanced · unknown · unknown resource · en - ### [Tracing the thoughts of a large language model](/security-and-red-teaming/safety-and-alignment-evals/tracing-the-thoughts-of-a-large-language-model/index.md) Creator not supplied; see original source. Anthropic interpretability research tracing internal mechanisms of Claude with attribution graphs. Advanced · unknown · unknown resource · en - ### [UK education generative-AI product safety standards](/security-and-red-teaming/safety-and-alignment-evals/uk-education-generative-ai-product-safety-standards/index.md) Creator not supplied; see original source. UK Department for Education generative AI product safety standards for AI tools used in education. Advanced · unknown · unknown resource · en - ### [A Mathematical Framework for Transformer Circuits](/security-and-red-teaming/safety-and-alignment-evals/a-mathematical-framework-for-transformer-circuits/index.md) Creator not supplied; see original source. Anthropic Transformer Circuits paper A Mathematical Framework for Transformer Circuits (2021). Expert · unknown · unknown resource · en - ### [AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents](/security-and-red-teaming/safety-and-alignment-evals/agentdojo-2/index.md) Creator not supplied; see original source. arXiv paper (2024) introducing AgentDojo, an environment for evaluating prompt injection attacks and defences on tool agents. Expert · unknown · unknown resource · en - ### [AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents](/security-and-red-teaming/safety-and-alignment-evals/agentharm-2/index.md) Creator not supplied; see original source. arXiv paper (2024) introducing AgentHarm, a benchmark measuring harmful task completion by LLM agents. Expert · unknown · unknown resource · en - ### [Amnesty: Xenophobic Machines (Dutch childcare benefits)](/security-and-red-teaming/safety-and-alignment-evals/amnesty-xenophobic-machines-dutch-childcare-benefits/index.md) Creator not supplied; see original source. Amnesty International report Xenophobic Machines on discriminatory algorithmic profiling in the Dutch childcare benefits scandal. Expert · unknown · unknown resource · en - ### [Defeating Prompt Injections by Design (CaMeL)](/security-and-red-teaming/safety-and-alignment-evals/camel-defeating-prompt-injections-by-design/index.md) Creator not supplied; see original source. arXiv paper (2025) by Google DeepMind introducing CaMeL, a capability-based design that defeats prompt injections. Expert · unknown · unknown resource · en - ### [Embrace The Red (Johann Rehberger)](/security-and-red-teaming/safety-and-alignment-evals/embrace-the-red-johann-rehberger/index.md) Creator not supplied; see original source. Embrace The Red, Johann Rehberger's blog documenting prompt-injection and data-exfiltration exploits in real AI agents and IDEs. Expert · unknown · unknown resource · en - ### [Emergent Misalignment](/security-and-red-teaming/safety-and-alignment-evals/emergent-misalignment/index.md) Creator not supplied; see original source. arXiv paper (2025) "Emergent Misalignment", showing narrow fine-tuning on insecure code produces broadly misaligned behaviour. Expert · unknown · unknown resource · en - ### [On the Biology of a Large Language Model](/security-and-red-teaming/safety-and-alignment-evals/on-the-biology-of-a-large-language-model/index.md) Creator not supplied; see original source. Anthropic Transformer Circuits paper On the Biology of a Large Language Model: attribution graphs on Claude 3.5 Haiku. Expert · unknown · unknown resource · en - ### [The Instruction Hierarchy](/security-and-red-teaming/safety-and-alignment-evals/the-instruction-hierarchy/index.md) Creator not supplied; see original source. arXiv paper (2024) by OpenAI introducing the instruction hierarchy to train models to prioritise privileged instructions. Expert · unknown · unknown resource · en - ### [Universal adversarial attacks (GCG)](/security-and-red-teaming/safety-and-alignment-evals/universal-adversarial-attacks-gcg/index.md) Creator not supplied; see original source. arXiv paper (2023) introducing GCG, universal transferable adversarial suffixes that jailbreak aligned LLMs. Expert · unknown · unknown resource · en [Previous page](/security-and-red-teaming/safety-and-alignment-evals/page/3/index.md) [Plain-text Markdown copy](/security-and-red-teaming/safety-and-alignment-evals/page/4/index.md)