Announcing Safety Research Grants - Thinking Machines Lab

Funding for the community to strengthen the ecosystem's defenses.

We recently released Inkling and Inkling-Small with open weights, and explained how we are thinking about safety and openness. Today, we are launching Tinker grants of up to $50,000 in credits for safety research on open-weight models. Below are some directions we find promising. The list is non-exhaustive on purpose. If you’re working on a safety project that could be accelerated by additional Tinker credits, we want to hear from you. Making models safer to open Differential acceleration of defensive over offensive capability. Many capabilities are dual-use: the same cyber skills that let…

saved by

• Ratan Kaliani
• Lydia Nottingham
• Florent Tavernier
• Bhagyesh Kumar

related reading

• A Safe Path to Open Weights - Thinking Machines Labthinkingmachines.ai
• [2507.11473] Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safetyarxiv.org
• Inkling: Our Open-Weights Model - Thinking Machines Labthinkingmachines.ai
• Announcing Interactivity Research Grants - Thinking Machines Labthinkingmachines.ai
• Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
• Automated Weak-to-Strong Researcheralignment.anthropic.com
• Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
• AI Safety is Bottlenecked - Opportunity for Impact is Highbriandavies.substack.com
• AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org
• [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
• Ten AI safety projects I'd like people to work onthirdthing.ai
• Security incident disclosure — July 2026huggingface.co