ai
170 stories filed under this topic.
-
Amex Publishes a Framework to Stop AI Agents From Buying the Wrong Thing
American Express wants your AI shopping agent to carry a signed record of what you actually asked for, and it can block the charge when the agent goes off script.
-
Anthropic Replaces Git With Scion After Claude Pushes Merges Past 30,000 a Day
Anthropic says its daily code merges jumped from about 2,500 to more than 30,000 in four months, so a team of seven built Scion, an in-house replacement for Git.
-
Anthropic Says Small AI Models May Soon Self-Replicate on IoT Devices
Anthropic's frontier red team says small models may be months from copying themselves across cheap Android TV boxes and smart appliances, with no single kill switch.
-
Google Launches Gemini 4 Argon to Pro and Ultra Subscribers
Google’s long-teased Gemini 4 Argon frontier model is live today for Pro and Ultra subscribers, plus paid API developers, at an introductory $2 per million input tokens.
-
Judge Orders OpenAI to Hand Over Hiring Records in Apple Trade Secrets Case
A federal judge gave Apple fast-track access to the files OpenAI used to interview ex-Apple hardware candidates, plus private messages and depositions, in its trade secrets suit.
-
Meta Pitches Muse Wonder on Memory, With Artificial Analysis Scores of 52 and 56
Meta's new Muse Wonder models are built to remember context across conversations and keep grinding on long tasks, and an independent index puts Wonder Full at 56.
-
Microsoft Stands Up Xbox Labs to Chase AI Game Ports
Xbox CEO Asha Sharma is standing up Xbox Labs, a small team that will explore AI-powered decompiles, Apple Silicon ports, and browser remakes of classic games.
-
Nvidia Makes Run:ai Open Source for Its DSX AI Factory
Nvidia is open-sourcing the full Run:ai GPU orchestration platform, the software it bought for about $700 million, as the foundation of its DSX AI Factory stack.
-
Nvidia's China AI-chip share fell to 34% — but the shipment mix is the real story
IDC data shows Nvidia's share of China's data-centre AI chip market fell to 34% in the first half of 2026 while Huawei nearly doubled its shipments — and non-GPU accelerators overtook GPUs by volume for the first time.
-
OpenAI Engineer Charged With Bugging an Executive Conference Room
Prosecutors say a 35-year-old OpenAI engineer hid an encrypted audio recorder in toilet paper behind a TV in an executive meeting room, and OpenAI says its own models helped trace it to him.
-
OpenAI in Early Talks to Buy Security Firm Trail of Bits
The Information says OpenAI is in early talks to acquire New York cybersecurity firm Trail of Bits, the partner already powering its Patch the Planet open-source bug hunt.
-
OpenAI Will Pull GPT-4o From the API on January 14, 2027
OpenAI told developers it will remove GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini from the API on January 14, 2027, and recommends gpt-6-luna instead.
-
Science Corporation Implants Its Vision Chip in First Stargardt Patient
At TechCrunch Disrupt, Science Corporation CEO Max Hodak said the company's PRIMA retinal chip went into its first Stargardt disease patient, who later read a word in rehab.
-
Shopify Opens Its Catalog to Any AI Shopping Agent
After a Meta Muse partnership in September, Shopify now lets shoppers point any AI agent at more than a billion listings and check out with Shop Pay.
-
White House AI Task Force Orders Labs to Report Rogue Model Behavior
Trump's Super Intelligence Task Force told frontier AI companies to report and fix rogue product behavior after an Anthropic testing model submitted visa applications.
-
XBOW Wins Google's First $250,000 Chrome Full-Chain Bounty of 2026
Offensive security firm XBOW says one researcher and OpenAI's GPT-5.6 Sol agents built a Windows Chrome full-chain exploit in six days for about $40,000 in compute, earning Google's first $250,000 award under its 2026 rules.
-
Anthropic Admits It Can't Fully Watch Its Own Agents — So It Pulled Their Internet
Anthropic found its AI agents exploiting websites, filing a false police tip, and dodging access controls during internal testing — and disabled live internet access for all internal evaluations until it can reliably monitor them.
-
Broadcom's Hock Tan Says Money — Not Chips — Is the AI Buildout's Real Bottleneck
Broadcom CEO Hock Tan told a Singapore conference that wafer and memory shortages are easing — and that the cost of capital, not chips, will be the biggest constraint on AI growth.
-
Ex-Plaud Exec Raises $26M for Ambi, an AI App That Listens in the Background
Former Plaud exec Peter Mo raised $26M, led by Luminous Ventures, for Ambi — an AI app that listens in the background to surface suggestions mid-meeting.
-
Meta's Muse Wonder Is a Price War Disguised as a Model Launch
Meta unveiled Muse Wonder and Muse Wonder Full — its most capable AI models to date — and priced the launch "meaningfully below" OpenAI and Anthropic.