ABOUT US
FAR.AI http://FAR.AI is a non-profit AI research institute working to ensure advanced AI is safe and beneficial for everyone. Our mission is to facilitate breakthrough AI safety research, advance global understanding of AI risks and solutions, and foster a coordinated global response.
Since our founding in July 2022, we've grown to 50+ staff https://www.far.ai/about/team, published 40+ academic papers https://scholar.google.com/citations?user=FVJ24k8AAAAJ, and convened leading AI safety events https://far.ai/events/. Our work is recognized globally, with publications at premier venues such as NeurIPS, ICML including a Best Paper Honorable Mention in 2026 https://icml.cc/virtual/2026/oral/71065, and ICLR, and features in the Financial Times https://www.ft.com/content/175e5314-a7f7-4741-a786-273219f433a1, Nature News https://www.nature.com/articles/d41586-024-02218-7, Wired Magazine https://www.wired.com/story/jailbreaking-ai-models-google-anthropic-openai-spacexai/ and MIT Technology Review https://www.technologyreview.com/2020/02/28/905615/reinforcement-learning-adversarial-attack-gaming-ai-deepmind-alphazero-selfdriving-cars/. We conduct pre-deployment testing on behalf of frontier developers such as OpenAI, independent evaluations for governments including the EU AI Office https://www.far.ai/news/far-ai-selected-to-lead-eu-ai-act-cbrn-risk-consortium, and publish the AI Security Leaderboard https://leaderboard.far.ai/ based on our red-teaming expertise. We help steer and grow the AI safety field through developing https://arxiv.org/abs/2405.06624 research https://arxiv.org/abs/2506.20702 roadmaps https://www.researchgate.net/publication/396910034_Open_Technical_Problems_in_Open-Weight_AI_Model_Risk_Management with renowned researchers such as Yoshua Bengio; running FAR.Labs https://www.far.ai/programs/far-labs, an AI safety-focused co-working space in Berkeley housing 40+ members; and supporting the community through targeted grants https://www.far.ai/programs/grantmaking to technical researchers.
ABOUT THE APPLIED WHITE-BOX METHODS TEAM
The Applied White-Box Methods team develops, evaluates, and demonstrates methods that leverage model internals to improve the safety of AI systems. We work on diverse AI safety applications of white-box methods, from white-box control to evaluation awareness to shaping training dynamics to improve alignment.
Black-box methods, such as chain-of-thought monitoring, work well for now, but we are quickly entering a world where black-box interventions and monitoring are insufficient. Interpretability research is still often early-stage, curiosity-driven work without realistic evaluations on applications that matter. The team bridges the gap between exploration and deployment by stress-testing white-box methods on real-world (e.g. long-context agentic coding) tasks and using this feedback loop to enable the deployment of better white-box methods at frontier scale.
Our scope includes any method that uses model internals to understand, predict, or intervene on model behavior, not only what is conventionally called interpretability. Methods of current interest include natural language autoencoders and other activation explainers, activation oracles, steering, patching, and other activation-level interventions, influence functions and data attribution, and singular learning theory. We evaluate our methods against real baselines – strong black-box methods and activation probes – to be able to make an honest case that the methods are worth implementing, or conclude that simpler methods work better for now.
AI Research Automation. AI will soon automate most of the hill-climbing in the research process. We anticipate this by focusing our effort and judgement on defining realistic evaluations with Goodhart-resistant metrics. With the evaluation framework properly built, we can pour vast amounts of AI labor into method development and iteration without overfitting. We believe this is the way to…