
Guide Labs is a product-focused research company dedicated to building interpretable AI systems that humans can reliably understand, debug, and trust. With over 20 years of experience and more than 24 published research papers, the team aims to address the shortcomings of current AI models, which often produce unreliable explanations and are difficult to control. By engineering models that prioritize interpretability and reliability, Guide Labs seeks to create AI systems that facilitate meaningful interactions between domain experts and technology, ultimately enhancing the usability and trustworthiness of AI in various applications.

Guide Labs is a product-focused research company dedicated to building interpretable AI systems that humans can reliably understand, debug, and trust. With over 20 years of experience and more than 24 published research papers, the team aims to address the shortcomings of current AI models, which often produce unreliable explanations and are difficult to control. By engineering models that prioritize interpretability and reliability, Guide Labs seeks to create AI systems that facilitate meaningful interactions between domain experts and technology, ultimately enhancing the usability and trustworthiness of AI in various applications.
Headquarters: San Francisco, California
Focus: Interpretable and auditable AI / interpretable LLMs
Notable model: Steerling-8B (open-sourced 8B interpretable model)
Team size: Approximately 10 employees
Recent funding: Seed announced Nov 2024 (~$9–9.3M); pre-seed $500K (Apr 2024)
Interpretable and auditable AI, interpretable large language models
Artificial Intelligence
$500,000
Investors listed: Y Combinator, Pioneer Fund, Lombardstreet Ventures (Crunchbase)
$9,300,000
Company reports $9.3M seed closed Nov 2024; press reports $9.0M
“Backers include Initialized Capital, Tectonic Ventures, Lombardstreet Ventures, Pioneer Fund, Y Combinator, and E14 (logos shown on company site and listed in reporting)”