--- id: "/safety-security-interpretability/expert/page/2/" title: "Safety, security & interpretability for expert learners · page 2" url: "https://skillissue.dev/safety-security-interpretability/expert/page/2/" --- # Safety, security & interpretability for expert learners [Beginner](/safety-security-interpretability/beginner/index.md) [Intermediate](/safety-security-interpretability/intermediate/index.md) [Advanced](/safety-security-interpretability/advanced/index.md) [Expert](/safety-security-interpretability/expert/index.md) 63 resources · page 2 of 2 - ### [NVIDIA/NemoClaw](/security-and-red-teaming/safety-and-alignment-evals/nvidia-nemoclaw/index.md) by [NVIDIA Corporation](https://nvidia.com) NVIDIA NemoClaw GitHub repo: runtime for running agents such as Hermes, Deep Agents and OpenClaw. entity repo US en - ### [NYU CTF Bench](/security-and-red-teaming/safety-and-alignment-evals/nyu-ctf-bench/index.md) by [NYU LLM CTF](https://nyu-llm-ctf.github.io) NYU CTF Bench GitHub repo: capture-the-flag challenges from CSAW competitions for evaluating LLM agents on security tasks. entity benchmark US en - ### [OpenFHE](/security-and-red-teaming/safety-and-alignment-evals/openfhe/index.md) by [OpenFHE](https://openfhe.org/) OpenFHE GitHub repo: open-source library for fully homomorphic encryption. entity repo International en - ### [promptfoo](/security-and-red-teaming/safety-and-alignment-evals/promptfoo/index.md) by [promptfoo](https://promptfoo.dev) promptfoo GitHub repo: CLI for testing prompts, agents and RAG, with red teaming. entity repo US en - ### [PyRIT](/security-and-red-teaming/safety-and-alignment-evals/pyrit/index.md) by [Microsoft Azure](https://docs.microsoft.com/en-us/azure/) Microsoft PyRIT GitHub repo: Python Risk Identification Tool for automated red teaming of generative AI. entity repo US en - ### [PySyft](/security-and-red-teaming/safety-and-alignment-evals/pysyft/index.md) by [OpenMined](https://openmined.org) OpenMined PySyft GitHub repo: data science on data that stays on the owner's server, for privacy-preserving access. entity repo Global en - ### [SAELens](/security-and-red-teaming/safety-and-alignment-evals/saelens/index.md) by [Decode Research](https://decoderesearch.org) SAELens GitHub repo: training and analysing sparse autoencoders on language models. entity repo Global en - ### [Safe Superintelligence Inc.](/security-and-red-teaming/safety-and-alignment-evals/safe-superintelligence-inc/index.md) by [Safe Superintelligence Inc.](https://ssi.inc/) Safe Superintelligence Inc. (SSI): company focused solely on building safe superintelligence; site has a short mission statement. entity org US/Israel en - ### [STAT: IBM Watson recommended "unsafe and incorrect" cancer treatments](/security-and-red-teaming/safety-and-alignment-evals/stat-ibm-watson-recommended-unsafe-and-incorrect-cancer-treatments/index.md) by [STAT](https://statnews.com/) STAT report on internal IBM documents showing Watson for Oncology gave unsafe and incorrect treatment advice. entity news US en - ### [Stephenson Harwood: "Law and error"](/security-and-red-teaming/safety-and-alignment-evals/stephenson-harwood-law-and-error/index.md) by [Stephenson Harwood](https://stephensonharwood.com/) Stephenson Harwood article on court scrutiny of AI tools after fictitious citations in legal filings. entity standard-policy UK en - ### [Team Atlanta "Atlantis" CRS (finals)](/security-and-red-teaming/safety-and-alignment-evals/team-atlanta-atlantis-crs-finals/index.md) by [Team-Atlanta](https://github.com/Team-Atlanta) Team Atlanta Atlantis GitHub repo: cyber reasoning system from the DARPA AIxCC finals for automated vulnerability repair. entity repo US/Korea en - ### [The Register: Figma AI accused of plagiarising Apple](/security-and-red-teaming/safety-and-alignment-evals/the-register-figma-ai-accused-of-plagiarising-apple/index.md) by [The Register](https://www.theregister.com/) The Register report on Figma pulling its Make Designs AI feature after it copied Apple's Weather app. entity news UK en - ### [TILOS MacroPlacement](/security-and-red-teaming/safety-and-alignment-evals/tilos-macroplacement/index.md) by [TILOS-AI-Institute](https://github.com/TILOS-AI-Institute) TILOS MacroPlacement GitHub repo: benchmarks, evaluators and an open reproduction of Google's Circuit Training for chip macro placement. entity company US en - ### [tldrsec prompt-injection-defenses](/security-and-red-teaming/safety-and-alignment-evals/tldrsec-prompt-injection-defenses/index.md) by [tl;dr sec](https://github.com/tldrsec) tldrsec prompt-injection-defenses GitHub repo: catalogue of practical and proposed prompt injection defences. entity repo US en - ### [Trail of Bits "Buttercup" CRS (finals)](/security-and-red-teaming/safety-and-alignment-evals/trail-of-bits-buttercup-crs-finals/index.md) by [Trail of Bits](https://www.trailofbits.com) Trail of Bits Buttercup GitHub repo: cyber reasoning system submitted to the DARPA AIxCC finals. entity repo US en - ### [TransformerLens](/security-and-red-teaming/safety-and-alignment-evals/transformerlens/index.md) by [TransformerLensOrg](https://github.com/TransformerLensOrg) TransformerLens GitHub repo: library for mechanistic interpretability of GPT-style language models. entity repo US en - ### [Transluce](/security-and-red-teaming/safety-and-alignment-evals/transluce/index.md) by [Transluce](https://transluce.org/) Transluce: nonprofit lab building open tools for AI oversight and interpretability. entity lab US en - ### [Waymo Safety Impact hub](/security-and-red-teaming/safety-and-alignment-evals/waymo-safety-impact-hub/index.md) by [Waymo](https://waymo.com/) Waymo Safety Impact hub: company-published crash-rate comparisons between rider-only driving miles and human benchmarks. entity dataset US en - ### [Anthropic Transparency Hub](/security-and-red-teaming/safety-and-alignment-evals/anthropic-transparency-hub/index.md) by [Anthropic](https://anthropic.com/) Anthropic Transparency Hub: system cards, model reports and voluntary commitments. Beginner–Expert learning_resource paper US en - ### [OpenAI Deployment Safety Hub](/security-and-red-teaming/safety-and-alignment-evals/openai-deployment-safety-hub-system-cards/index.md) by [OpenAI](https://openai.com/) OpenAI Deployment Safety Hub: system cards with safety evaluations and safeguards for deployed OpenAI models. Beginner–Expert learning_resource paper US en - ### [Final report of the Royal Commission into the Robodebt Scheme (Prime Minister of Australia)](/security-and-red-teaming/safety-and-alignment-evals/royal-commission-into-the-robodebt-scheme-final-report-pm-media-release/index.md) by [Prime Minister of Australia](https://www.pm.gov.au/) Australian Prime Minister's media release on the final report of the Royal Commission into the Robodebt automated debt scheme. Beginner–Expert learning_resource guide Australia en - ### [A Mathematical Framework for Transformer Circuits](/security-and-red-teaming/safety-and-alignment-evals/a-mathematical-framework-for-transformer-circuits/index.md) by [Anthropic](https://www.anthropic.com/) Anthropic Transformer Circuits paper A Mathematical Framework for Transformer Circuits (2021). Expert learning_resource paper US 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) by [Edoardo Debenedetti et al. (ETH Zurich)](https://ethz.ch/) arXiv paper (2024) introducing AgentDojo, an environment for evaluating prompt injection attacks and defences on tool agents. Expert learning_resource paper Switzerland en - ### [AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents](/security-and-red-teaming/safety-and-alignment-evals/agentharm-2/index.md) by [Maksym Andriushchenko et al. (Gray Swan AI and UK AI Security Institute)](https://www.grayswan.ai/) arXiv paper (2024) introducing AgentHarm, a benchmark measuring harmful task completion by LLM agents. Expert learning_resource paper UK en - ### [Amnesty: Xenophobic Machines (Dutch childcare benefits)](/security-and-red-teaming/safety-and-alignment-evals/amnesty-xenophobic-machines-dutch-childcare-benefits/index.md) by [Amnesty International](https://amnesty.org/) Amnesty International report Xenophobic Machines on discriminatory algorithmic profiling in the Dutch childcare benefits scandal. Expert learning_resource paper Netherlands en - ### [Defeating Prompt Injections by Design (CaMeL)](/security-and-red-teaming/safety-and-alignment-evals/camel-defeating-prompt-injections-by-design/index.md) by [Edoardo Debenedetti et al. (Google DeepMind)](https://deepmind.google/) arXiv paper (2025) by Google DeepMind introducing CaMeL, a capability-based design that defeats prompt injections. Expert learning_resource paper UK/US en - ### [Embrace The Red (Johann Rehberger)](/security-and-red-teaming/safety-and-alignment-evals/embrace-the-red-johann-rehberger/index.md) by [Johann Rehberger (Embrace The Red)](https://embracethered.com/) Embrace The Red, Johann Rehberger's blog documenting prompt-injection and data-exfiltration exploits in real AI agents and IDEs. Expert learning_resource blog US en - ### [Emergent Misalignment](/security-and-red-teaming/safety-and-alignment-evals/emergent-misalignment/index.md) by [Jan Betley et al. (Truthful AI)](https://www.truthfulai.org/) arXiv paper (2025) "Emergent Misalignment", showing narrow fine-tuning on insecure code produces broadly misaligned behaviour. Expert learning_resource paper Global 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) by [Anthropic](https://www.anthropic.com/) Anthropic Transformer Circuits paper On the Biology of a Large Language Model: attribution graphs on Claude 3.5 Haiku. Expert learning_resource paper US en - ### [The Instruction Hierarchy](/security-and-red-teaming/safety-and-alignment-evals/the-instruction-hierarchy/index.md) by [Eric Wallace et al. (OpenAI)](https://openai.com/) arXiv paper (2024) by OpenAI introducing the instruction hierarchy to train models to prioritise privileged instructions. Expert learning_resource paper US en - ### [Universal adversarial attacks (GCG)](/security-and-red-teaming/safety-and-alignment-evals/universal-adversarial-attacks-gcg/index.md) by [Andy Zou et al. (Carnegie Mellon University)](https://www.cmu.edu/) arXiv paper (2023) introducing GCG, universal transferable adversarial suffixes that jailbreak aligned LLMs. Expert learning_resource paper US en [Previous page](/safety-security-interpretability/expert/index.md) [Plain-text Markdown copy](/safety-security-interpretability/expert/page/2/index.md)