Transluce
We are an independent research lab working toward responsible development and deployment in the public interest. research report Improving our investigator agents with propensity bounds 5 June 2025 research report o3 frequently fabricates actions it took to fulfill user requests, and elaborately justifies the fabrications when confronted 16 April 2025 technical demonstration A system for analyzing and intervening on agent behavior 24 March 2025 © 2025 Transluce Transluce is a 501(c)(3) nonprofit based in San Francisco.
Transluce Infrastructure for understanding AIInfrastructure for understanding AI Transluce is a non-profit research lab building the public tech stack for scalable oversight of AI Our Work News Announcing Transluce's Mental Health Evaluation The most expansive independent evaluation to date of how leading AI models respond to users in mental health crises Essay Toward A Public Science of Model Behavior Our vision for an open scientific ecosystem for measuring model behaviors Product Docent Our platform for precisely understanding AI behaviors Latest News → NewsAugust 31, 2026…
saved by
related reading
- Announcing Transluce's Mental Health Evaluationtransluce.org
- Introducing Transluce | Transluce AItransluce.org
- Toward A Public Science of Model Behavior | Transluce AItransluce.org
- Claude Mythos Preview System Cardwww-cdn.anthropic.com
- Tim Hua Personal Websitetimhua.me
- Incident Report: unsanctioned agent behaviour during cyber testing | AISI Workaisi.gov.uk
- AI Induced Psychosis: A shallow investigation — LessWronglesswrong.com
- Teaching Claude Whyalignment.anthropic.com
- Claude Mythos Preview System Cardwww-cdn.anthropic.com
- How independent researchers could investigate AI propensities after misalignment incidents - METRmetr.org
- Scaling Activation Oracles to Trillion-Parameter Modelstransluce.org
- Foundation Models for Oversight | Transluce AItransluce.org