TLDRocket
Sign in

Benchmarking

148 summarised stories about Benchmarking, each linking back to the original source. Browse all topics →

Monday, 13 April 2026

Import AI 453: Breaking AI agents; MirrorCode; and ten views on gradual disempowerment

Import AI 3 months ago

Researchers demonstrated that Claude Opus 4.6 successfully reimplemented a 16,000-line bioinformatics program (gotree) by observing only its command-line interface and test cases, without access to the original source code. The task would typically require a human engineer 2–17 weeks of work, and performance improved with increased computational resources during inference. This capability suggests AI systems can already autonomously complete weeks-long coding tasks, raising questions about the pace of AI progress in software development.