A recent evaluation suggests that GPT-5.2 is performing at an expert level in scientific peer review. In a study involving 45 scientists and 469 hours, AI reviews were found to be competitive with top human reviewers on 82 papers. However, the AI still has weaknesses, indicating that a hybrid approach of AI and human collaboration is optimal for peer review. AI
IMPACT AI models are becoming competitive with human experts in complex tasks like scientific peer review, suggesting potential for increased efficiency and quality in research.
RANK_REASON The cluster describes a study evaluating an AI model's performance on a specific task (peer review), which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — sigmoid.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →