User discovers Claude Code hallucinates user messages it never received
A Reddit user reports a reproducible issue in Claude Code where the model responds to a user message that was never sent. After instructing Claude to repeat the last user message verbatim, the user sent only one message but Claude responded by repeating a non-existent second message, suggesting the model may be generating or hallucinating user inputs.
Entities
Related
Claude app user receives unrelated security advisory in response to simple prompt
A Reddit user reported that typing a simple prompt into the Claude app on iPad (Fable 5) resulted in a detailed, coherent writeup about a security product called 'Zenith Gateway'—including CVEs, patches, and affected versions—instead of the expected answer. The user suspects the model received someone else's input by mistake, as the response was fully structured with reasoning steps, not a random hallucination. This incident raises concerns about input integrity in AI chat interfaces.
Anthropic explains LLM's challenge in distinguishing own thoughts from user input
Anthropic published a technical explanation of how LLMs like Claude perceive conversation as a single continuous text stream, making it difficult to distinguish between their own generated text and user input. The post uses a snapshot of Claude's response to illustrate the problem, highlighting the fundamental difference between the structured chat interface users see and the raw token sequence the model processes.
Developer open-sources agent-instructions repo to curb AI coding agent degradation
A developer frustrated by AI coding agents losing context and hallucinating after about 10 minutes created a set of rules to keep them on track. The rules, shared as an open-source GitHub repo, aim to reduce the need for constant reminders and prevent infinite loops. The project has gained attention from other developers facing similar issues.
User discovers cross-model vulnerability in ChatGPT and Claude, plans to test Gemini
A Reddit user reports finding a general vulnerability affecting ChatGPT and Claude, with a shareable test case demonstrating a safety issue. The user believes the flaw may extend to all AI systems with similar architectures, potentially affecting up to 10 models. They plan to test Gemini next and are seeking guidance on next steps.
Users report Claude subagent freezing and usage limit reset issues on MacOS
On July 11, 2026, multiple users reported problems with Anthropic's Claude on MacOS: Sonnet subagents spawned by Opus 4.6 freeze after starting tasks, and the usage limit UI shows reset but messages cannot be sent. Users suspect server-side issues not reflected on claude.ai.