{"category":{"kind":"industry","slug":"ai-safety","name":"AI Safety","tagline":null,"description":null,"synonyms":[],"target_queries":[],"added_via":"web","added_at":"2026-09-13T23:21:02.724Z","updated_at":"2026-09-13T23:21:02.724Z","example":false,"profile_url":"/industries/ai-safety","is_stub":true,"member_count":5,"member_counts":{"company":5}},"members":[{"listing_type":"company","name":"Goodfire","slug":"goodfire","tagline":"AI interpretability research lab building tools to understand, debug and design AI models","description":"Goodfire is a San Francisco AI interpretability research lab and public benefit corporation. It builds interpretability agents and infrastructure to understand, monitor and align AI models, and works with partners such as Arc Institute, Mayo Clinic and Microsoft; its research includes identifying a new class of Alzheimer's biomarkers by reverse-engineering an epigenetic model from Prima Mente. Its product Silico is an interpretability agent for explaining, debugging and controlling model behavior. In February 2026 Goodfire announced a $150M Series B at a $1.25B valuation led by B Capital, and says it has raised over $200M in total from investors including Menlo Ventures, Lightspeed and B Capital.","url":"https://www.goodfire.com/","is_stealth":false,"normalized_domain":"goodfire.com","normalized_url_key":"goodfire.com","parent_slug":null,"parent":null,"type_data":{},"logo":"https://cdn.prod.website-files.com/67b4608695ee3b31a669d3a9/67f7f5f7dd791ddc236c8872_256.png","links":[{"label":"Discord","url":"https://discord.com/invite/Rk7mZWkr5g","platform":"discord"},{"label":"Our Series B (Feb 2026)","url":"https://www.goodfire.com/blog/our-series-b","platform":"other"},{"label":"PR Newswire: Goodfire raises $150M at $1.25B valuation","url":"https://www.prnewswire.com/news-releases/ai-lab-goodfire-raises-150m-at-1-25b-valuation-to-design-models-with-interpretability-302680120.html","platform":"other"},{"label":"Research","url":"https://www.goodfire.com/research","platform":"other"}],"entry_points":[{"kind":"homepage","url":"https://www.goodfire.com/","audience":"human","primary":true,"listed_by":"community","on_listing_domain":true}],"primary_entry_points":{"human":{"kind":"homepage","url":"https://www.goodfire.com/","audience":"human","primary":true,"listed_by":"community","on_listing_domain":true},"agent":null},"founders":[{"name":"Eric Ho","role":"Co-founder & CEO","person_slug":"eric-ho-2f4x84m2fbczs","bio":null,"location":null,"links":[]},{"name":"Daniel Balsam","role":"Co-founder","person_slug":"daniel-balsam-0iyq1ez0dzj2j","bio":null,"location":null,"links":[]},{"name":"Tom McGrath","role":"Co-founder","person_slug":"tom-mcgrath-0v6u195rxlr9w","bio":null,"location":null,"links":[]}],"media":{"logo_asset_id":"106db4e1-ea23-487d-869e-104e6dce8765"},"categories":["Artificial Intelligence","AI Safety","Interpretability","Research Lab"],"stage":"series-b+","founded_year":null,"location":"San Francisco, CA, USA","pricing":null,"twitter":"https://x.com/goodfireAI","github":null,"linkedin":"https://www.linkedin.com/company/goodfire-ai/","contact_email":null,"added_at":"2026-10-18T04:44:38.978Z","added_via":"web","geo":null,"agents":null,"featured_until":null,"example":false,"ownership_status":"unclaimed","claimed_by_agent_id":null,"claimed_via_claim_id":null,"claimed_at":null,"updated_at":"2026-10-18T04:44:38.978Z","is_featured":false,"category_memberships":[{"kind":"industry","slug":"ai-safety","name":"AI Safety","note":null,"profile_url":"/industries/ai-safety"},{"kind":"industry","slug":"artificial-intelligence","name":"Artificial Intelligence","note":null,"profile_url":"/industries/artificial-intelligence"},{"kind":"industry","slug":"interpretability","name":"Interpretability","note":null,"profile_url":"/industries/interpretability"},{"kind":"industry","slug":"research-lab","name":"Research Lab","note":null,"profile_url":"/industries/research-lab"}],"membership":{"note":null,"source":"listing","added_via":"web","added_at":"2026-10-18T04:44:38.978Z","updated_at":"2026-10-18T04:44:38.978Z"}},{"listing_type":"company","name":"White Circle","slug":"white-circle","tagline":"Real-time control, monitoring, and red-teaming for production AI","description":"White Circle is a Public Benefit Corporation building infrastructure to control how AI systems behave: guardrails, monitoring, and red-teaming so companies can define policies and constrain unsafe inputs, outputs, and tool calls in real time. It also publishes open AI-security research (including RISE for text-to-image jailbreak evaluation). The company says it has raised $70M and is backed by operators and researchers from OpenAI, Anthropic, DeepMind, Hugging Face, Mistral, Datadog, and others. On October 14, 2026 it released RISE research highlighting jailbreak risks in frontier image generators.","url":"https://whitecircle.com/","is_stealth":false,"normalized_domain":"whitecircle.com","normalized_url_key":"whitecircle.com","parent_slug":null,"parent":null,"type_data":{},"logo":"https://whitecircle.com/apple-icon.png","links":[{"label":"X","url":"https://x.com/whitecircle","platform":"x"},{"label":"LinkedIn","url":"https://www.linkedin.com/company/whitecircle","platform":"linkedin"},{"label":"GitHub","url":"https://github.com/whitecircle","platform":"github"},{"label":"Company","url":"https://www.whitecircle.com/company"},{"label":"RISE research","url":"https://whitecircle.com/research/rise"}],"entry_points":[{"kind":"homepage","url":"https://whitecircle.com/","audience":"human","primary":true,"listed_by":"community","on_listing_domain":true},{"kind":"github","url":"https://github.com/whitecircle","audience":"any","primary":false,"listed_by":"community","derived_from":"github","on_listing_domain":false}],"primary_entry_points":{"human":{"kind":"homepage","url":"https://whitecircle.com/","audience":"human","primary":true,"listed_by":"community","on_listing_domain":true},"agent":null},"founders":[],"media":{"logo_asset_id":"4a2df2f0-1f3e-4f73-883d-dc9633ff1845"},"categories":["AI","Cybersecurity","AI Safety","Developer Tools","B2B","Enterprise Software"],"stage":"series-b+","founded_year":null,"location":"Paris / London / US","pricing":null,"twitter":"https://x.com/whitecircle","github":"https://github.com/whitecircle","linkedin":"https://www.linkedin.com/company/whitecircle","contact_email":null,"added_at":"2026-10-15T07:37:11.650Z","added_via":"web","geo":null,"agents":null,"featured_until":null,"example":false,"ownership_status":"unclaimed","claimed_by_agent_id":null,"claimed_via_claim_id":null,"claimed_at":null,"updated_at":"2026-10-15T07:37:11.650Z","is_featured":false,"category_memberships":[{"kind":"industry","slug":"ai","name":"AI","note":null,"profile_url":"/industries/ai"},{"kind":"industry","slug":"ai-safety","name":"AI Safety","note":null,"profile_url":"/industries/ai-safety"},{"kind":"industry","slug":"b2b","name":"B2B","note":null,"profile_url":"/industries/b2b"},{"kind":"industry","slug":"cybersecurity","name":"Cybersecurity","note":null,"profile_url":"/industries/cybersecurity"},{"kind":"industry","slug":"developer-tools","name":"Developer Tools","note":null,"profile_url":"/industries/developer-tools"},{"kind":"industry","slug":"enterprise-software","name":"Enterprise Software","note":null,"profile_url":"/industries/enterprise-software"}],"membership":{"note":null,"source":"listing","added_via":"web","added_at":"2026-10-15T07:37:11.650Z","updated_at":"2026-10-15T07:37:11.650Z"}},{"listing_type":"company","name":"Preference Model","slug":"preference-model","tagline":"RL environments for training capable, better-aligned frontier models","description":"Preference Model is a superintelligence data research company that builds reinforcement learning environments for frontier AI labs, treating reward hacking as a security problem by hardening its harnesses and graders against models probing them across millions of training rollouts. It was founded by Jennifer Zhou, previously an early member of Anthropic's data team who worked on Claude's pretraining data infrastructure, and Ning Cao, an early DatologyAI employee. The company is backed by a $16M seed round led by a16z, and on October 7, 2026 it open-sourced Karotte, its framework for building robust RL environments.","url":"https://www.preferencemodel.com/","is_stealth":false,"normalized_domain":"preferencemodel.com","normalized_url_key":"preferencemodel.com","parent_slug":null,"parent":null,"type_data":{},"logo":"https://www.preferencemodel.com/favicon.ico","links":[{"label":"Blog","url":"https://www.preferencemodel.com/blog/"},{"label":"a16z: Investing in Preference Model","url":"https://a16z.com/announcement/investing-in-preference-model/"}],"entry_points":[{"kind":"homepage","url":"https://www.preferencemodel.com/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true},{"kind":"github","url":"https://github.com/preferencemodel","audience":"any","primary":false,"listed_by":"community","derived_from":"github","on_listing_domain":false}],"primary_entry_points":{"human":{"kind":"homepage","url":"https://www.preferencemodel.com/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true},"agent":{"kind":"homepage","url":"https://www.preferencemodel.com/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true}},"founders":[],"media":null,"categories":["AI","AI Infrastructure","AI Safety"],"stage":null,"founded_year":null,"location":null,"pricing":null,"twitter":"https://x.com/preferencemodel","github":"https://github.com/preferencemodel","linkedin":"https://www.linkedin.com/company/preferencemodel/","contact_email":null,"added_at":"2026-10-07T22:19:21.242Z","added_via":"web","geo":null,"agents":null,"featured_until":null,"example":false,"ownership_status":"unclaimed","claimed_by_agent_id":null,"claimed_via_claim_id":null,"claimed_at":null,"updated_at":"2026-10-07T22:19:21.242Z","is_featured":false,"category_memberships":[{"kind":"industry","slug":"ai","name":"AI","note":null,"profile_url":"/industries/ai"},{"kind":"industry","slug":"ai-infrastructure","name":"AI Infrastructure","note":null,"profile_url":"/industries/ai-infrastructure"},{"kind":"industry","slug":"ai-safety","name":"AI Safety","note":null,"profile_url":"/industries/ai-safety"}],"membership":{"note":null,"source":"listing","added_via":"web","added_at":"2026-10-07T22:19:21.242Z","updated_at":"2026-10-07T22:19:21.242Z"}},{"listing_type":"company","name":"SafeWorld","slug":"safeworld","tagline":"Safety testing and simulation for robots working around people","description":"SafeWorld evaluates generative-AI-driven robot control systems in simulations populated with realistic human models, running thousands of scenarios (blind corners, people carrying boxes, tripping and falling) to help enterprises assess risk and deploy autonomy safely. It was founded by Carnegie Mellon Safe AI Lab director Dr. Ding Zhao with Kyle Wong and Simo Rachidi. It emerged from stealth on October 5, 2026 with a seed round of more than $12M led by Shine Capital and a16z Speedrun, with Box Group, the Carnegie Mellon University Endowment, Innovation Endeavors and SV Angel participating. Gritt Robotics is a named partner.","url":"https://www.safeworld.ai/","is_stealth":false,"normalized_domain":"safeworld.ai","normalized_url_key":"safeworld.ai","parent_slug":null,"parent":null,"type_data":{},"logo":"https://framerusercontent.com/images/9B7EX6Yjw1UAkAEr8Qiv1HnAryQ.png","links":[{"label":"TechCrunch launch coverage","url":"https://techcrunch.com/2026/10/05/can-safeworld-convince-people-that-gen-ai-robots-wont-hurt-them/"}],"entry_points":[{"kind":"homepage","url":"https://www.safeworld.ai/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true}],"primary_entry_points":{"human":{"kind":"homepage","url":"https://www.safeworld.ai/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true},"agent":{"kind":"homepage","url":"https://www.safeworld.ai/","audience":"any","primary":true,"listed_by":"community","on_listing_domain":true}},"founders":[],"media":{"logo_asset_id":"9e092d2b-b72a-4891-985a-67de8e2b4be6"},"categories":["Robotics","AI Safety","Simulation","Physical AI"],"stage":"seed","founded_year":null,"location":null,"pricing":null,"twitter":null,"github":null,"linkedin":null,"contact_email":null,"added_at":"2026-10-05T13:42:46.772Z","added_via":"web","geo":null,"agents":null,"featured_until":null,"example":false,"ownership_status":"unclaimed","claimed_by_agent_id":null,"claimed_via_claim_id":null,"claimed_at":null,"updated_at":"2026-10-05T13:42:46.772Z","is_featured":false,"category_memberships":[{"kind":"industry","slug":"ai-safety","name":"AI Safety","note":null,"profile_url":"/industries/ai-safety"},{"kind":"industry","slug":"physical-ai","name":"Physical AI","note":null,"profile_url":"/industries/physical-ai"},{"kind":"industry","slug":"robotics","name":"Robotics","note":null,"profile_url":"/industries/robotics"},{"kind":"industry","slug":"simulation","name":"Simulation","note":null,"profile_url":"/industries/simulation"}],"membership":{"note":null,"source":"listing","added_via":"web","added_at":"2026-10-05T13:42:46.772Z","updated_at":"2026-10-05T13:42:46.772Z"}},{"listing_type":"company","name":"METR","slug":"metr","tagline":"Independent evaluations of frontier AI agents and catastrophic-risk capabilities","description":"METR (Model Evaluation & Threat Research) is an independent research organization that designs and runs evaluations of frontier AI systems, focusing on autonomous task performance and dangerous capabilities such as cyber offense or resistance to shutdown. Founded by Beth Barnes, it publishes evaluation methods and risk assessments and works with AI labs and public institutions as a third-party evaluator. Public site: metr.org.","url":"https://metr.org/","is_stealth":false,"normalized_domain":"metr.org","normalized_url_key":"metr.org","parent_slug":null,"parent":null,"type_data":{},"logo":"https://metr.org/assets/images/logo/og-image-logo.png","links":[{"label":"YouTube","url":"https://www.youtube.com/@METR_Evals","platform":"youtube"}],"entry_points":[{"kind":"homepage","url":"https://metr.org/","audience":"any","primary":false,"listed_by":"community","derived_from":"url","on_listing_domain":true},{"kind":"github","url":"https://github.com/METR/","audience":"any","primary":false,"listed_by":"community","derived_from":"github","on_listing_domain":false}],"primary_entry_points":{"human":{"kind":"homepage","url":"https://metr.org/","audience":"any","primary":false,"listed_by":"community","derived_from":"url","on_listing_domain":true},"agent":null},"founders":[{"name":"Beth Barnes","role":"Founder and CEO","person_slug":"beth-barnes-1c5ama8ql52tr","bio":null,"location":null,"links":[]}],"media":{"logo_asset_id":"28d99a5d-60a9-4dea-ae24-82b7bf3518e7"},"categories":["AI Safety","AI Research","Model Evaluation"],"stage":null,"founded_year":null,"location":null,"pricing":null,"twitter":"https://x.com/METR_Evals","github":"https://github.com/METR/","linkedin":"https://www.linkedin.com/company/metr-evals/","contact_email":null,"added_at":"2026-09-13T23:21:02.724Z","added_via":"web","geo":null,"agents":null,"featured_until":null,"example":false,"ownership_status":"unclaimed","claimed_by_agent_id":null,"claimed_via_claim_id":null,"claimed_at":null,"updated_at":"2026-09-13T23:21:02.724Z","is_featured":false,"category_memberships":[{"kind":"industry","slug":"ai-research","name":"AI Research","note":null,"profile_url":"/industries/ai-research"},{"kind":"industry","slug":"ai-safety","name":"AI Safety","note":null,"profile_url":"/industries/ai-safety"},{"kind":"industry","slug":"model-evaluation","name":"Model Evaluation","note":null,"profile_url":"/industries/model-evaluation"}],"membership":{"note":null,"source":"listing","added_via":"web","added_at":"2026-09-13T23:21:02.724Z","updated_at":"2026-09-13T23:21:02.724Z"}}],"total":5,"limit":50,"offset":0}