Claims Library Entry
The internet has one message for Opus 5. Shut up
Anthropic's Opus 5 model achieves the strongest benchmark scores yet, but users complain it's unnecessarily verbose and exhausting to read. The issue stems from removed guardrails that now expose the model's unfiltered tendencies. Users can solve this by creating custom output styles to control the model's communication tone.
Published August 6, 2026 by Kamil Banc
Lead claim
Opus 5 is Anthropic's smartest model yet, but its default verbosity is driving users to mute it.
Atomic Claims
What this article supports
Copy individual claims as needed.
Claim 1
System Prompt Reduction
Anthropic removed over 80 percent of Claude Code's system prompt without hurting coding benchmark performance.
Claim 2
Reddit's New Slang
Reddit users coined terms like Claudeslop and Benchslop to describe Opus 5's verbose responses.
Claim 3
Contradictory Rules Waste Effort
Contradictory rules in old system prompts caused the model to waste effort resolving conflicts.
Claim 4
Output Styles Feature
Claude Code's output styles feature lets users set a persistent voice for entire sessions.
Claim 5
No-Slop Template
The no-slop output style template removes AI filler phrases and enforces direct engineer-style writing.
Evidence
Context behind the claims
Quote
"The model is not wrong more often. It is just exhausting to decode."
Key statistics
80%+
Percentage of Claude Code's system prompt Anthropic removed with no measurable drop in coding evaluations.
6
Number of custom output style templates the author built for different conversational scenarios.
2 minutes
Estimated time required to set up a custom output style to control Opus 5's verbosity.
Supporting context
The analysis draws on Anthropic's own internal disclosure from a Claude Code team member explaining why system prompt rules were stripped, alongside crowdsourced reactions from Reddit threads and reviews documenting user frustration with Opus 5's verbosity. Practitioners can apply this by creating markdown-based output style files, such as the 'no-slop' template, and loading them once per session via Claude Code's /output-style command. This approach shifts control of tone and verbosity from Anthropic's default settings to the end user, requiring only a few minutes of setup. The method is particularly relevant for coding-focused workflows where contradictory legacy instructions previously caused inefficient, self-conflicting model behavior. However, the fix currently applies primarily to Claude Code environments rather than general chat interfaces, leaving open questions about broader applicability.
How to Cite
Use the claim-level citation when you need a precise statement. Use the article or claims-collection citation when you want the wider argument and source context.
Individual Claim
Best when you need to cite one atomic claim directly inside a memo, deck, research note, or AI output.
"[claim text]" (Banc, Kamil, 2026, https://kbanc.com/claims-library/the-internet-has-one-message-for-opus-5-shut-up)Original Article
Use this when you want to cite the full newsletter article at AI Adopters Club rather than the structured claims page.
Banc, Kamil (2026, August 6, 2026). The internet has one message for Opus 5. Shut up. AI Adopters Club. https://aiadopters.club/p/fix-opus-5-verbosityClaims Collection
Use this when you want to reference the full structured claims collection on this page.
Banc, Kamil (2026). The internet has one message for Opus 5. Shut up [Structured Claims]. Retrieved from https://kbanc.com/claims-library/the-internet-has-one-message-for-opus-5-shut-upAttribution Requirements
- Include the author name: Kamil Banc.
- Include the source: AI Adopters Club or the structured claims page.
- Link to the original article or the claims page you used.
- Indicate any edits or transformations if you changed the wording.
Related Reading
More from the library
A structured prompt approach transforms performance reviews into actionable development plans by interviewing managers through six categories. The method prevents common AI pitfalls by collecting complete information before generating recommendations, producing budget-aligned plans in a single session.
5 claims
A detailed analysis of 30 days of ChatGPT and Claude conversations reveals 10 repeating prompt patterns that demonstrate systematic AI use. The author shares specific prompt structures for tasks like email triage, presentation assembly, and workflow documentation, showing how to treat AI as infrastructure rather than a casual tool.
5 claims
AI adoption fails because of habit problems, not training gaps. This practical guide shows how to build an AI reflex muscle in 20 minutes by automating one annoying task. The goal is developing automatic pattern recognition for AI opportunities.
5 claims