China’s LineShine tops the June 2026 TOP500 supercomputer list, though mixed-precision results leave El Capitan stronger on ...

AI benchmark cheating has been theorized as an inevitable consequence of training capable optimizers against fixed metrics. With OpenAI's GPT-5.6 Sol, the theory arrived in full view. The nonprofit ...

DXOMARK evaluates the camera performance of seven leading smartglasses, comparing image quality outdoors, indoors, and in low light against the iPhone 13 selfie camera.

Local AI inference at 32B-parameter quality, no cloud API required: University of Waterloo researchers released PAW on July 2 ...

Mass General Brigham, Mount Sinai and Baptist Health South Florida leaders explain how they vet clinical AI tools — and why the OpenEvidence dispute is mostly noise.

Meta’s new AI research vice president, Dawn Song, says AI agents must prove they can complete useful real-world work.

Local AI inference at 32B-parameter quality, no cloud API required: University of Waterloo researchers released PAW on July 2, 2026, a system that compiles any natural-language task spec into a 23MB ...

Cybersecurity has become one of the cleaner growth stories in software because AI is making the threat surface larger.

The South African Reserve Bank plans to end the use of prime as a benchmark interest rate, but there are massive operational ...

DSpark can make decoding faster, but acceptance quality still determines how much speed the system actually realizes.

Award-winning "scan-to-cup" wellness platform debuts at CHI Innovate 2026 (2–4 July) and powers Singapore's first AI-driven ...

Two Chinese AI tools now match Anthropic's Mythos in cybersecurity vulnerability detection. Both are freely available, ...