AI still needs humans

AI coding tools are increasingly used to generate software from prompts, with claims that most new code can be AI-generated and then reviewed by engineers. Warnings from engineers behind widely used AI agents describe “vibe slop,” where people skip key software development work such as design, judgment, testing, ownership, and deep system understanding. The concern is that AI coding can flood industry with buggy and potentially dangerous software when used at industrial scale. The response is to rethink assumptions rather than reject AI coding. AI coding can be powerful like power tools, but it can also enable unskilled or careless users to make larger mistakes with greater confidence. Nearly correct code can still be very wrong and carries real costs.

"If you truly want to avoid drowning in 'AI slop,' you need experienced engineers who can supervise the observability, testing, and review of all that AI-generated code."

"The complaint is that too many people are using AI to skip the parts of software development that actually matter: design, judgment, testing, ownership, and deep understanding of the system being changed."

"When people who helped build the tools used by millions start warning that those same tools can produce buggy, potentially dangerous software at industrial scale, it's probably time to rethink some of the assumptions fueling the AI wave."

"AI coding is powerful in roughly the same way power tools are powerful. They help skilled people do more, faster. They also help unskilled or careless people make bigger mistakes with greater confidence. That's the enterprise AI story in miniature."

Read at InfoWorld