AI Makes All Code 'Write-Only': What That Means for Developers and the Rest of Us

Key points

• Engineer Phillip Mortimer argued at QCon London 2025 that AI turns all code into "write-only" material: produced faster than any human can read or check it.
• Tests, not the code itself, now serve as the real documentation for what software does.
• Mortimer recommends AI agents to review code automatically and to fix live problems by watching a system's alerts.
• Because AI translates plain instructions into any language, developers no longer need to master a specific programming language's syntax.
• Solo thinking, research cited by Mortimer suggests, produces better ideas than group brainstorming sessions.

"Write-only code" is a decades-old programmer's insult for software so tangled that even its author can't read it back. At QCon London, engineer Phillip Mortimer argued that AI has made every codebase write-only by default.

That sounds alarming. Here's what it actually means.

So what is the problem, exactly?

AI coding tools can produce hundreds of lines of working software in seconds, far quicker than any human team can check it. Volume alone makes line-by-line review impossible. The code works, but nobody fully understands it.

Mortimer's answer isn't to slow down. It's to stop pretending humans can keep up. Two principles follow.

First: tests define the behaviour. A test is a small automated check that feeds a program a known input and confirms the right output. When code is too dense to follow, those input-output pairs become the only readable record of what the software is supposed to do.

Second: code is disposable. If something breaks and the AI-written code is hard to unpick, the faster move is often to delete it and rewrite from scratch. Keep the tests. Discard the implementation.

Can AI review its own code?

Yes, and Mortimer says it works better than you'd expect. An AI agent, software that carries out multi-step tasks on its own, can slot into the same pipeline developers already use to check each other's work. A different prompt and a different context seems enough to catch real bugs, even in code another AI just wrote.

He also described "de-slopify" agents: tools that scan a codebase for repeated code, unparameterised tests or unused feature flags, and clean them up without human prompting.

Beyond that, Mortimer envisions software that heals itself. An AI agent watches the live system's alerts, spots patterns, and raises a proposed fix for the engineering team to approve. Our earlier story on Anthropic's Claude Code showed a similar instinct at work: in testing, autonomous action caught harmful steps far more reliably than human oversight did.

What does this mean for developers' jobs?

Mortimer's answer: the job shifts toward creativity. Writing the actual lines of code is something AI handles. What remains is knowing what to build and why.

Developers need uninterrupted stretches to think. Brainstorming sessions, which most workplaces love, have been shown to produce fewer good ideas than individuals working alone. Constraints help too: a tight deadline or a limited budget forces more creative thinking than a blank canvas.

For non-developers, the shift is concrete. The software running your bank or your hospital records is increasingly written and patched by AI. The humans overseeing it are moving from writing every line to setting goals and checking outcomes. That's a different kind of oversight, and the industry's still working out how to do it well.

Common questions

Does write-only AI code mean software is less safe?

Not automatically, but it does change where the risk sits. Fewer human eyes on individual lines means automated testing and AI review tools carry more of the safety burden than before.

Do developers need to learn new programming languages now?

Mortimer argues the opposite: because AI translates plain instructions into any language, developers can work across many languages without mastering each one's syntax from scratch.