An independent lab with a narrow question.
Golden Toad Research asks whether the numbers used to approve an AI system describe how it behaves in practice. A model can score well on overlap, accuracy, or a refusal check and still fail on the quantity a clinician or operator relies on. We look for those gaps, measure them carefully, and report what we find plainly.
James Irwin
Principal researcher
LinkedInJames Irwin graduated from Texas Tech University in 2026 with a B.S. in Computer Science and a B.A. in Mathematics. His prior work includes leading medical imaging and machine learning research in the Mathematics Department's Lung Cancer Research Lab, AI security research in the Computer Science Department's Cybersecurity Research Lab, and EEG signal analysis in the Electrical Engineering Department.
- Lung Cancer Research Lab, MathematicsTexas Tech
- Cybersecurity Research Lab, Computer ScienceTexas Tech
- EEG research, Electrical EngineeringTexas Tech
- B.S. Computer ScienceTexas Tech, 2026
- B.A. MathematicsTexas Tech, 2026
- Study abroadUVA · Università di Siena
We are excited to work and publish with researchers from a wide range of backgrounds.
Medicine, neuroscience, computer science, mathematics, engineering, or a field we have not touched yet. If you have a question about how AI systems are measured, we look forward to hearing from you.
ji.globalcl@gmail.com jamirwin@ttu.edu LinkedIn