Mira Murati’s Thinking Machines Lab Makes The Technical Case For Human-Centered AI Built On Customizable Model Weights
Mira Murati's post-OpenAI move: human-centered AI isn't philosophy—it's a technical architecture built on LoRA fine-tuning and model ownership.
Why it matters
Murati's Thinking Machines Lab is staking out a contrarian technical position on AI alignment and governance: decentralized model weights and user ownership as infrastructure, not afterthought. This signals a emerging fault line in how AI labs approach control, safety, and deployment post-2026.
The key facts
10 to knowMira Murati founded Thinking Machines Lab (post-OpenAI departure)
Published essay: 'The Future Worth Building Is Human'
Frames human participation, model ownership, decentralized alignment as technical challenges
References LoRA fine-tuning and customizable model weights as implementation vehicle
Positions teams training/owning their own weights as core technical approach
UNVERIFIED: Publication date is July 2026 (future-dated; verify source credibility)
Mira Murati founded Thinking Machines Lab
Core thesis: human participation, model ownership, decentralized alignment as technical challenges
LoRA fine-tuning and customizable model weights positioned as key enabler
Frames interaction models and team-level model control as alternative to centralized LLM dependency
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: Thinking Machines Lab published "The Future Worth Building Is Human." The essay frames human participation, model ownership, and decentralized alignment as technical challenges. It ties them to interaction models and Tinker's LoRA fine-tuning, where teams train and keep their own model weights.