Timaeus is an AI safety research nonprofit dedicated to making fundamental breakthroughs in technical AI alignment using deep ideas from mathematics and the sciences. The organization applies Singular Learning Theory (SLT), combining insights from algebraic geometry and statistical physics, to develop interpretability tools that reveal how internal structure forms during neural network training. Their flagship research agenda, developmental interpretability, uses measures like the Local Learning Coefficient to detect and understand phase transitions in model training, with the goal of building scalable tools for evaluating, interpreting, and aligning AI systems.
Timaeus is an AI safety research nonprofit dedicated to making fundamental breakthroughs in technical AI alignment using deep ideas from mathematics and the sciences. The organization applies Singular Learning Theory (SLT), combining insights from algebraic geometry and statistical physics, to develop interpretability tools that reveal how internal structure forms during neural network training. Their flagship research agenda, developmental interpretability, uses measures like the Local Learning Coefficient to detect and understand phase transitions in model training, with the goal of building scalable tools for evaluating, interpreting, and aligning AI systems.
Funding Details
- Annual Budget
- $2,500,000
- Monthly Burn Rate
- $208,333
- Current Runway
- -
- Funding Goal
- -
- Funding Raised to Date
- -
- Fiscal Sponsor
- Ashgro
Theory of Change
Timaeus believes that Singular Learning Theory provides the mathematical foundation needed to rigorously understand how neural networks develop internal structure during training, including when and how they acquire capabilities, values, and potentially dangerous behaviors. By detecting and understanding phase transitions in training through measures like the Local Learning Coefficient, Timaeus aims to build scalable evaluation and interpretability tools that can identify critical developmental moments before dangerous properties emerge. This would enable alignment interventions at the training level, ensuring models develop in ways that reflect human values. Their approach bridges pure mathematics with empirical machine learning, creating a scientific framework where alignment properties can be formally stated, measured, and verified rather than merely hoped for.
Grants Received
from Open Philanthropy
from Open Philanthropy
from Survival and Flourishing Fund
from Long-Term Future Fund
from Survival and Flourishing Fund
from Survival and Flourishing Fund
Projects– no linked projects
People– no linked people
Discussion
Sign in to join the discussion.
Key risk: The main risk is that SLT-derived signals (e.g., LLC) may fail to generalize to safety-relevant phenomena in real frontier systems, leaving their elegant math and promising mid-scale demos without actionable interventions or lab adoption, especially given limited access to compute and proprietary models.
Details
- Last Updated
- Apr 2, 2026, 10:10 PM UTC
- Created
- Mar 18, 2026, 11:18 PM UTC
Case for funding: Timaeus is uniquely positioned to turn Singular Learning Theory into practical, scalable developmental interpretability (LLC, spectroscopy, patterning, devinterp) that can flag dangerous phase transitions during training—an approach already shaping top lab agendas and likely to create training-time levers for frontier models.