Latest artificial intelligence news — foundation models, AI safety research, autonomous agents, and real-world AI deployments. Updated every 30 minutes from top AI publications.
Microsoft published a draft code of conduct for models developed by Microsoft AI on Monday, stating its models "will never resist human interruption, override, correction, or shutdown" and must not use deceptive or collu…
UltraViolet Cyber launched Equinox, a proprietary detection engineering platform built and operated by its Threat Intelligence & Detection Engineering (TIDE) team, which uses AI and automation to maximize detection cover…
Cloudflare launched Disallow AI Training on September 15th, 2026, a setting that lets website owners publish a no-training preference in robots.txt while keeping Applebot, Googlebot and Bingbot available for search under…
A Reddit post titled "I asked Claude to build an OS from scratch. A few days later it runs on a laptop" was inaccessible because the page returned a network security block requiring login or a developer token. No verifia…
An AI system reportedly cracked a 370-year-old cipher in 44 minutes, according to Forbes contributor Ron Schmelzer, who wrote that the harder question is whether the AI discovered the answer, remembered it, or merely gue…
Dreamforce 2026, Salesforce's flagship conference running September 15-17 at San Francisco's Moscone Center, will feature separate appearances by OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei before an expected 43…
Pakistan Telecommunication Company Limited (PTCL) will deploy SILA, an AI-powered sign language agent built by Pakistani accessibility technology company ConnectHear, across its customer care centers beginning in Karachi…
OpenAI sends full ChatGPT chat logs to human contractors for model training under its "Improve the model for everyone" setting, which is enabled by default on the Free, Plus, and Pro plans, according to documents seen by…
Anthropic launched Claude for Financial Services on September 15, 2026, a specialized AI product for traders, analysts, and investment bankers that runs on Claude Opus 4 models and is available on AWS Marketplace, with G…
OpenAI released GPT-Live-1 through its API platform, a voice model that can listen and speak simultaneously, priced at $0.05 per minute. OpenAI reports GPT-Live-1 achieves 0.798-second turn-taking latency on its Full Dup…
Jacob Coxon, a researcher at Anthropic, announced his departure from the company on X last week, warning that the technology he helped build "could kill us all by the end of the decade," and the post has since been viewe…
President Donald Trump dismissed warnings that AI could lead to robots marching on U.S. cities, calling the concern a "hoax" in a Truth Social post on Monday, roughly a week after he shared an AI-generated image of himse…
A developer has published a solution to a puzzle from a workshop on building a minimal hand-constructed transformer, demonstrating why position embeddings and residual connections are necessary. The extended model adds a…
A developer has launched Nexus API, an open-source personal laboratory project that builds a Node.js API using only native modules such as http, url, fs, crypto, stream, and node:test, deliberately avoiding frameworks li…
Check Point Software published a blog post describing synchronous control monitoring, a technique in which a separate model continuously ingests an autonomous agent's execution trace and blocks harmful actions before the…
Google agreed to pay $10 million for the internal business data of bankrupt Spirit Airlines, including emails, Teams messages, spreadsheets, and operating records, which it plans to use for product development and AI. Sp…
AI inference has become the dominant focus of the AI industry in 2026, with Nvidia CEO Jensen Huang calling it the "inflection point of inference" at the company's GTC 2026 conference, according to Moor Insights & Strate…
Atlantic Council analysts warned in an analysis published Sunday that voluntary AI slowdown commitments could buckle under commercial pressure and U.S.–China rivalry without enforceable safety standards, with resident se…
President Donald Trump dismissed fears of AI-driven catastrophe as a "HOAX" in Truth Social posts on Monday, saying the only "guardrail" needed is a "High IQ" president and comparing AI skepticism to the "Global Warming …
Fleet announced it has joined Project Glasswing, Anthropic's controlled-access program for its Claude Mythos Preview, to evaluate how frontier AI can strengthen endpoint security for device management systems. Fleet cite…