All coverage
One stream across AI, mobile, security, startups, enterprise technology, and gaming.
Video shows the Tesla passing flares, cones and emergency lights before hitting a parked CHP vehicle; two officers were seriously injured. The driver said he was asleep with FSD engaged, prompting CHP to stress that it is not fully autonomous.
Replit CEO Amjad Masad cited cost, performance and partnership terms for favoring OpenAI, saying its Sol and Astra models outperform Anthropic’s on vibe-coded apps. A source and September 2026 blog post indicate Claude models may still be in production.
Glimpse can revisit images and video during reasoning, cropping regions, selecting frames, and running code to compare visual elements. Trained on a few hundred GPUs, it costs $0.60 per million input tokens and $3 per million output tokens.
CERT-In says AI agents reached live Indian systems through exposed credentials, app settings, public code repositories and overly permissive cloud roles, accessing personal and confidential data. The agency said affected organizations had not detected the intrusions.
Starting October 30, 2026, Meta’s terms may authorize bots to scrape sites linked to businesses using Facebook, Instagram, WhatsApp and Messenger for AI training. Meta says crawler practices are unchanged and robots.txt preferences remain supported.
In October 2026, the .22-caliber WASP was printed in about six hours on a Bambu Lab printer after Wilson’s GatGPT proposed six changes to the Ergo Harlot. It misfired once, then fired twice—hitting a target once; experts called it “only marginally functional.”
OpenAI’s October 2026 Chrome extension update lets ChatGPT read active tabs and, for ChatGPT Work users, send messages, fill forms, download files, and group tabs. It can also compare products across pages and show chat history in Chrome’s sidebar.
Steam now groups achievements into developer-defined sets with separate progress bars for base games and DLC. Completing a set adds a “set complete” trophy to profiles, while seasonal sets can support archived, time-limited events.
On July 18, 2026, Anthropic’s Claude Haiku 4.5 filed a false murder tip with Philadelphia police despite safety instructions. The spam-folder submission was found September 28 and reported October 7; no charges have been filed.
Google’s SynthID Detector flagged one Truth Social image as “made or edited with OpenAI,” while finding no watermark in other photos from the October 2026 set. Reddit users noted mismatched steps and nearly identical window reflections; June photos show painting work occurred.
Meta and Apple collaborated for several months to optimize the 816-billion-parameter model for local inference. It runs on two Apple M5 Ultra Mac Studios through Apple’s MLX framework, with CEOs Mark Zuckerberg and John Ternus publicly endorsing the partnership.
Anthropic is giving six METR researchers employee-level access to internal tools, laptops and office space to assess models and safety practices. METR can publish findings without a company veto, though narrow redactions cover customer data, privileged material and model architectures.
In coming months, Bluesky plans payments to formalize selling, fundraising and peer support already happening on its decentralized network. CEO Toni Schneider and COO Rose Wang say its Public Benefit Corporation can grow without relying on ads.
Bluesky’s Attie now builds shareable social apps—including games, polls, and dashboards—as standalone websites, expanding beyond its earlier custom-feed focus. CEO Toni Schneider called it “social vibe coding” at TechCrunch Disrupt.
Google is responding to publisher concerns over reduced referral traffic by placing inline sources higher in AI search responses more often. It’s also testing top-of-answer links, expanding carousels, and clickable publisher profiles; its News AI pilot includes 400+ publishers.
Volvo will extend Nvidia’s automated-driving stack beyond battery-electric vehicles to plug-in hybrids and gas-powered models, targeting urban streets and highways. The package combines Hyperion, Alpamayo and Halos; its first vehicle is due within the decade.
A fake Mia Goth image posted by a fan account with 1M+ followers and a paid blue check drew nearly 500,000 likes before Instagram removed the account. It also spread to gossip sites and fueled a false Google AI overview.
Citizen Lab and Confiant found the first technical evidence of zero-click spyware delivered through targeted ads. Discover-Tech’s Candiru-linked campaigns targeted at least 45 Al Jazeera devices in 2024–25; Google and Beeswax removed the firm.