Claude Opus 4.7 draws backlash over no-think mode

Anthropic pitches Opus 4.7 as a model with adaptive thinking and stronger long-running workflows, so this backlash is really about product feel versus benchmark claims.

• –If users perceive “instant answer” behavior as less thoughtful, that can erode trust even when the model scores better on coding evals.
• –The criticism highlights a split between code-first agent workflows and broader assistant use cases like analysis, planning, and roleplay.
• –Visible reasoning cues matter as a UX contract; removing them may save latency but makes the model feel less capable on ambiguous tasks.
• –If Anthropic is optimizing Opus 4.7 for agentic throughput, some users will treat it as a downgrade for high-context conversational work.