Researchers at UCLA found GPT-3 solved 80% of reasoning problems correctly compared with 60% of humans
ChatGPT can solve problems at a level that matches or surpasses an undergraduate student, according to a new study.
Researchers found that the GPT-3 large language model that underpins the chatbot performed better than US college undergraduates when asked to solve reasoning problems that appear on intelligence tests or exams such as the American college admission test, the SAT.
More Stories
UK Biobank is safely sharing health data to drive medical research | Letter
Revealed: Chinese researchers can access half a million UK GP records
Nasa’s oldest astronaut celebrates 70th birthday while hurtling back to Earth – video