Cursor shows AI agents capable of shoddy code at scale

"A week ago, Cursor CEO Michael Truell celebrated what sounded like a remarkable event. "We built a browser with GPT-5.2 in Cursor," he said in a social media post. "It ran uninterrupted for one week." This browser, he said, consisted of three million lines of code across thousands of files. "The rendering engine is from-scratch in Rust with HTML parsing, CSS cascade, layout, text shaping, paint, and a custom JS VM.""

"But by and large, developers aren't convinced Cursor has made a breakthrough. Jason Gorman, managing director of Codemanship, a UK-based software development consultancy, argues it's proof that agentic AI scales to produce broken software. Oliver Medhurst, a software engineer and former Mozillan who participates in Ecma International's TC39 standards group, concurs. Asked whether there's anything more here than a demonstration that AI agents can produce large projects of not very high quality code, he said that's a good summary of the project."

""It is impressive that it does somewhat work and that it can (poorly) manage a codebase of that size but I would say that is what is impressive," Medhurst told The Register in an email. "Cursor said it was just a demonstration and I think it is fair to call it as such, but it is definitely not a good browser engine, objectively. Another point is that it is incredibly bloated. Ladybird and Servo do much more in much less lines of code (Ladybird and Servo repos are both ~1M).""

Cursor published a browser prototype claimed to be built with GPT-5.2 that reportedly ran uninterrupted for one week. The prototype is described as three million lines across thousands of files with a from-scratch Rust rendering engine handling HTML parsing, CSS cascade, layout, text shaping, paint, and a custom JavaScript VM. Some developers compiled the code after fixes or revised build instructions, but many developers remained unconvinced this is a breakthrough. Critics argue agentic AI can scale to produce large but broken software, cite high failure rates, and call the project bloated and far from WebKit/Chromium parity.

Read at Theregister