Top LLMs performed poorly on a high-level history test, a new paper has found.
AI isn’t very good at history, new paper finds | TechCrunch
All Versions
Jan-20
Ai
Left Variant
Score: 0.20
Recent research reveals that leading language models, the pride of the tech industry, struggle to grasp historical facts, underscoring the limits of technology's role in education and the importance of human teachers.
From TechCrunch
Jan-20
Ai
Right Variant
Score: 0.80
A new study indicates that even the most advanced AI systems falter on simple historical knowledge tests, raising questions about the programming biases inherent in Silicon Valley's creations and the reliability of education technology.
From TechCrunch