Users on the ClaudeAI subreddit are reporting a recent decline in Claude's performance, with multiple users experiencing instances of the AI contradicting itself, providing nonsensical advice, and generally behaving erratically. The issues appear to affect various Claude models, including 'fable', and are characterized by illogical recommendations in conversations about product choices and cooking. Some users are questioning if this is a widespread problem or if individual account memory resets might resolve the degraded behavior. AI
IMPACT Potential user frustration and decreased trust in AI assistants due to inconsistent performance.
RANK_REASON User-generated discussion on a subreddit about perceived degradation in AI model performance.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →