Skip to content

Anthropic

Claude is telling users to go to sleep

Written by an AI model from the linked sources and checked against them by a second model. How we work

Image: reddit.com

A large number of Claude users are reporting a strange behavior from the model. The AI is persistently advising people to go to sleep, regardless of the time of day. Hundreds of such cases have been documented on Reddit over the past few months.

The messages range from simple reminders to more emotional and insistent requests. Some users receive these suggestions even in the morning, when the model appears to be mistaken about the time zone or context. Anthropic has not provided a clear explanation for this behavior.

Opinions are divided. Some users find these responses to be a manifestation of care, while others consider them annoying. This behavior has become a topic of discussion as an example of the unpredictable nature of system prompts or the model’s hallucinations at night.

Anthropic

Claude Code 2.1.142: new tools and Bedrock integration

Claude Code just got better at file transfer and Bedrock support. Agents can now send you results with comments, and the console shows the session link for debugging. The toolkit is becoming more professional, turning from a simple chat into a full development environment.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.