Researchers have created a benchmark called MirrorCode to test AI models’ ability to reimplement complex software without access to the original source code. The results show that AI systems are surprisingly capable at certain coding tasks, with one model successfully reimplanting a 16,000-line bioinformatics toolkit. This has significant implications for the field of AI, as it demonstrates the ability of AI systems to generate functional equivalents of existing software. Why it matters - for some tasks, AI is already as good as a full-time sophisticated employee, autonomously completing complex tasks that would take humans weeks to accomplish.