Lenz
Independent, multi-model fact-checking API for AI workflows
398 followers
Independent, multi-model fact-checking API for AI workflows
398 followers
Lenz is an AI fact-checking API for products that cannot afford to hallucinate. It extracts verifiable claims from any text, then checks each one: searching independent sources, running multi-model debate, and routing through a review panel — returning a scored verdict with every source, argument, and step visible. Most AI tools give you one model's best guess from memory. Lenz ensures no single model's blind spots drive the conclusion. Available as API and MCP. Try it free at lenz.io/ph
This is the 2nd launch from Lenz. View more
Lenz
Launched this week
Lenz is an AI fact-checking API for products that cannot afford to hallucinate. It extracts verifiable claims from any text, then checks each one: searching independent sources, running multi-model debate, and routing through a review panel — returning a scored verdict with every source, argument, and step visible. Most AI tools give you one model's best guess from memory. Lenz ensures no single model's blind spots drive the conclusion. Available as API and MCP. Try it free at lenz.io/ph
Free Options
Launch tags:Artificial Intelligence
Launch Team / Built With
Paint the Cameras Dead
This feels especially important right now. We are living in a world where AI is becoming one of the first places people go to check what is true, yet your research shows that frontier models disagree on 63% of real-world fact checks.
That is such a powerful reminder that a confident AI answer is not the same thing as a verified answer.
Really love what you’re building with Lenz. The need for trustworthy verification is only going to grow as AI becomes more embedded in how we work, learn, and make decisions.
At this point, I’m just happy to have an AI that occasionally says, “Let me check that” instead of confidently making things up 😂
Lenz
@bogomep Indeed. In our experience, the models' self-reported level of confidence in their answers is mostly noise. They aren't trained to self-assess. Even worse, their incentives (to keep the user happy and engaged) push them to show high confidence when asked. On a 1-10 scale, in 76% of answers, the models report confidence of 9 or 10. And the correlation between confidence level and how much they agree with each other is not too strong either: https://lenz.io/research/llm-disagreement#confidence