# A developer emailed Claude Code's creator about AI slop. Boris Cherny wrote back. > Source: > Published: 2026-09-12 09:47:01+00:00 Developers are worried about AI. The creator of Claude Code has a response. On Thursday, Anthropic's [Boris Cherny](https://www.businessinsider.com/claude-code-boris-cherny-better-way-measure-ai-success-dashboards-2026-7) posted an email a developer wrote to him on X. The email was titled: "What to do about slop?" Cherny wrote that he receives emails like this every day and that he tries to respond to as many as possible. It's no surprise that developers are reaching out to Cherny with their worries. AI has [radically changed coding](https://www.businessinsider.com/great-coding-reset-ai-software-engineering-2026-7), shifting job descriptions and turning engineers into reviewers. Some engineers love the ease; others call it [lonely and tedious](https://www.businessinsider.com/openai-software-engineer-leaves-ai-coding-lonely-draining-2026-8?gift-id=3i_EGPSd5lynrZh6THvCuA). In the email, the unnamed developer wrote that the shift to AI coding was causing "friction." He broke it down into two views and asked Cherny to broker the dispute. The first view was a more restrained use of AI tools: We create code similar to before, accelerated by AI. We may not review all of it, but it should be reviewable. The person submitting the code should be able to explain it. It should be built in such a way to make it as easy to maintain — or easier — than before, by human agents. The second view was more permissive. This was the vibe coding position: We treat the code as a black box. We don't worry about it. We just check the output. The developer wrote that many people were incapable of doing option 1 — the more restrained approach — but worried about [getting laid off](https://www.businessinsider.com/list-companies-replacing-human-employees-with-ai-layoffs-workforce-reductions). "A lot of people are just forcing it," they wrote. Cherny posted his response in full. As for the two options the developer provided, Cherny said there was room for both. On the "don't worry" approach: Prototypes and other throw-away code can be treated as totally black box. If you're going to throw it away anyway, and if the blast radius of it breaking is low, it doesn't need to be perfect. And, on cautious reviewing: Production code written by Claude should have a higher bar than if it was written by a human. At Anthropic, we have many guardrails in place to make sure this is happening: lots of lint rules, lots of tests, Claude-driven end to end tests, Claude-powered fuzzers running daily, automated code reviews and security reviews, automated code refactoring, and so on. Without these, you can end up with a mess that is hard to maintain down the line. Luckily, the model makes it increasingly easy to do these well — run a few daily routines, use Claude Code Review, etc. The developer's job is to "hold the bar on code quality," Cherny wrote. AI leaders use different terms for this role, such as "[mini CEO](https://www.businessinsider.com/ai-agent-managers-vercel-guillermo-rauch-2026-3)." It's the developer's job to act like a manager, stopping [bad code](https://www.businessinsider.com/ai-writing-all-startup-code-thats-creating-a-new-problem-2026-6) in its tracks. If all else fails, Cherny said the developer could be more hands-on with Claude or ask it to rewrite the codebase. "Or, wait for the next model," he wrote.