AI coding benchmark MirrorCode published its full results June 26, showing Claude Opus 4.7 autonomously rebuilt a 60,000-line interpreter and scored 56% overall — completing tasks that take human ...

Bigger has defined AI from day one. New data says task-specific small models beat frontier LLMs on accuracy, cost and speed — and save money.

The Eleventh Conference on Machine Translation (WMT26) has moved into its active evaluation phase, with test data releases and submission windows now opening across several of the conference’s shared ...

Python’s lead narrows again, C holds the runner-up spot, C++ returns to third, and SQL climbs back above R in June’s top 10 ...

Look to these key metrics and benchmarks to evaluate the performance, capability, reliability, and safety of your AI models ...

Software developers working with command-line tools and large codebases now have a new option from Microsoft: ...

DeepSWE puts GPT-5.5 atop the AI coding leaderboard while raising new questions about Claude Opus, SWE-Bench Pro, and benchmark leakage.

ChatGPT, Claude, Grok, Gemini and other AI models display systematic religious bias, according to scientific research from ...

Programming languages shape how software, apps, and websites are built, making them one of the most important skills in the modern digital world. With industries shifting toward automation, AI tools, ...

While much attention regarding AI has been focused on developers using it to code, the impact of AI on software development goes far beyond code creation tools. Armando Solar-Lezama, Distinguished ...

One-off tests don’t measure AI’s true impact. We’re better off shifting to more human-centered, context-specific methods. For decades, artificial intelligence has been evaluated through the question ...

Early in the Covid-19 pandemic, the governor of New Jersey made an unusual admission: He’d run out of COBOL developers. The state’s unemployment insurance systems were written in the 60-year-old ...