Base Labs Launches Open-Weight AI Safety Partnership With Hugging Face and Goodfire
· Technology · TechCrunch
Baseten launched a new safety infrastructure standard alongside its Base Labs research arm on Wednesday, partnering with Hugging Face and Goodfire to build monitoring infrastructure for open-weight models. The initiative targets the growing issue of abliteration, a technique used to remove safeguards from open-source AI models, with Hugging Face currently hosting over 6,000 such models. Base Labs will develop and publish methods for training and monitoring open models to ensure transparency is built directly into deployment. Goodfire brings its model interpretability platform to the partnership to help explain how AI models make decisions. Baseten recently raised a $1.5 billion Series F in June, while Goodfire secured a $150 million Series B led by B Capital earlier this year.
Why it matters
The partnership creates standardized safety and monitoring infrastructure for open-weight AI models, directly impacting developers and enterprises deploying open-source artificial intelligence systems.
Read the original report — TechCrunch
Join us on Telegram
Breaking news the moment it lands. At 10,000 members we ship the Android app.