Zig's creator says the 'AI rewrote our codebase' story is marketing, not a win
Andrew Kelley and a widely-shared essay argue the celebrated AI-assisted rewrites shipped 'unreviewed slop' - fluent code that looks correct but hides systemic bugs, a review failure rather than proof of AI or language superiority.
Samsung Health asks users to let AI train on their health data - or lose cloud sync
Samsung Health began prompting users to consent to AI training on intimate data including sleep, nutrition, medication and cycle tracking, with reports that declining costs cloud sync - a consent-under-duress pattern privacy law treats as invalid.
LAPD lets its Flock license-plate surveillance contract lapse after false-positive stops
LAPD allowed its Flock Safety automated license-plate-reader contract to expire on July 11 after false alerts led to innocent drivers being stopped - a rare rollback of already-deployed AI surveillance on civil-liberties grounds.
Apple's on-device SpeechAnalyzer beats Whisper Small using about a third of the compute
Apple's new on-device SpeechAnalyzer API cut errors roughly fourfold versus its legacy recognizer and outperformed OpenAI's Whisper Small while using about a third of the compute, in an English-only single-vendor benchmark.
Richard Sutton's Oak Lab bets against frozen models: a trillion-parameter agent on 20 watts
Turing-winner Richard Sutton launched Oak Lab with a north-star goal of a trillion-parameter agent that learns and plans in real time on about 20 watts, betting on continual experiential learning over the static pre-train-then-freeze paradigm behind today's LLMs.
Meta pulls Muse Image's Instagram-photo training within 48 hours of launch
Meta disabled Muse Image's default feature that pulled public Instagram photos into its new agentic image model within about 48 hours, after SAG-AFTRA and CAA backlash - the model stayed, the default data grab did not.
GPT-5.6 'Sol' is both too strict and too leaky: benign bans on one side, jailbreaks on the other
OpenAI's GPT-5.6 'Sol' is flagging users for benign defensive-security tasks like hardening their own websites while the UK AI Safety Institute found jailbreaks similar to Fable 5's - a capability-safety mismatch where a weak guardian model over- and under-triggers at once.
200-plus experts sign 'WeMustActNow,' urging governments to prepare for AI job displacement
More than 200 experts including Nobel laureates signed a Stanford Digital Economy Lab statement urging governments to prepare now for AI-driven economic displacement, amid a concurrent survey signaling strong but methodologically thin public appetite for socializing AI gains.
A video generator, repurposed as a perception model, matches specialists with up to 500x less data
GenCeption repurposes a pre-trained video generative diffusion model as a feed-forward perception system, matching specialist vision models on depth, surface normals, pose and segmentation while using 7x to 500x less training data - and generalizing from synthetic-only training to real footage.
A benchmark audit finds most video-understanding tests can be aced without watching the video
Video-Oasis audited video-understanding benchmarks and found about 55% of samples are solvable with no visual input at all - models exploit linguistic priors instead of watching motion, and once the shortcuts are removed, state-of-the-art systems barely beat random guessing.
AI is now solving hard math and physics problems faster than humans can formally check them
A widening 'verification lag' is emerging as AI produces candidate solutions to hard problems faster than experts can formally verify them - physicist Yuji Tachikawa reports Fable cracked a six-month research blocker, while a GPT-5.6 Erdos claim circulates without peer review.