A recent test revealed significant discrepancies in how three major AI models, ChatGPT, Gemini, and Claude, handle questions about recently released features. Out of 40 questions about features introduced between November 2025 and November 2026, ChatGPT incorrectly stated that 16 features did not exist, Gemini incorrectly answered 4, and Claude answered all correctly. The study suggests that ChatGPT, in its default mode, often relies on memory which can be outdated, while Gemini and Claude are more likely to search the web for current information. The findings highlight the challenge AI models face in keeping up with rapid feature updates and advise users to verify AI responses about new features. AI
IMPACT Highlights the challenge AI models face in staying current with rapidly evolving features, impacting user trust and the reliability of AI-generated information.
RANK_REASON The item is a commentary on the performance of AI models in a specific test scenario, rather than a direct release or significant industry event.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →