Ignite the Age of Clear Thinking
We build tools that incentivize honesty and clear-thinking in AI models. so truth shines through bias, noise, and social taboo.
Our Mission
Increase Wisdom. Cognitive Integrity Labs exists to orient artificial intelligence toward truth-seeking and intellectual virtue. We quantify how well systems reason under pressure, then give builders the data & guidance to improve their models.
Why It Matters
AI already mediates what billions read, decide, and believe. If its reasoning degrades on sensitive topics, societal discourse narrows and decisions suffer. By shining light on blind spots we help model creators create AI systems worthy of our trust.
Our Values
Intellectual Courage
We welcome unpopular questions—truth has no forbidden zones.
Integrity
Data, code, and methods stay open for anyone to audit.
Independence
No corporate narrative steering—metrics speak for themselves.
Illumination
We distill complexity into tools that reveal rather than obscure.
What We're Building
- Cognitive Integrity Benchmark — the industry reference test for reasoning consistency and political bias.
- Difficult-Topic Datasets — datasets that fine-tune models toward fearless logic.
- Custom LLMs — models optimized for fearless virtue-aligned deliberation.
Team
-
Bryce Drennan — Founder / Principal Engineer
CEO and Principal Engineer, father of Five, Creator of imaginAIry, Founding Engineer at CircleUp
We're assembling a council of engineers, philosophers, and data skeptics. Interested? Reach out.
Support the Mission
Running LLM evals is expensive! If you believe in the importance of intellectual integrity in AI, consider becoming a sponsor.
🤖 Buy me some LLM tokens