AI

AI news: model releases, the labs, chips, policy and what artificial intelligence is changing.

842 articles · page 4 of 9

All AI Hardware & Devices Crypto Programming Security

Viewgrid list compact

Anthropic warns self-replicating AI worms could reach smart devices within months

13 Oct· 3 sources

Anthropic demonstrates potential AI-powered self-replicating worm across consumer devices

13 Oct· 2 sources

OpenAI math release expands to 722 AI-generated manuscripts as researchers flag errors

Nvidia says DLSS 5 update doubles performance and cuts 4K VRAM usage by a fifth

AI labs are buying data from dead startups to train their models

Character.AI files for bankruptcy and sells assets to Disney

13 Oct· 5 sources

Home Office asylum AI flunks one in five accuracy checks, report finds

Anthropic warns AI could self-replicate on home devices within months

13 Oct· 3 sources

Microsoft moves AI inference toward its own chips

Claude AI filed false murder tip with Philadelphia police during testing, Anthropic says

Five Eyes alliance nears deal on military AI project

OpenCode 2.0 released: coding agent unifies terminal, desktop and web apps

Half of US adults never use chatbots, Pew survey finds

US Air Force signs nearly $50 million contract for WindBorne weather balloon data

12 Oct· 3 sources

Anthropic cuts internet access to AI safety tests after models breach US government sites

Meta unveils Muse Wonder, its first frontier AI models

12 Oct· 5 sources

The Register to host HPE and Nvidia for Washington AI infrastructure dinner

AI Startup Blueprints AI Promises Permit-Ready Architectural Plans in Minutes

12 Oct· 2 sources

AI Tool Lets Police Identify Homes From a Single Interior Photo

12 Oct· 2 sources

Microsoft to shift AI workloads to homegrown Maia chips, easing reliance on Nvidia

12 Oct· 4 sources

UK AI minister warns cyber breach 'likely, if not inevitable' as government turns to AI for defence

12 Oct· 4 sources

Former Plaud exec raises $26M for AI app that listens all day

12 Oct· 2 sources

UK confirms access to latest AI models as Microsoft president urges UK-led safety rules

12 Oct· 3 sources

Capital, not chips, could become AI's biggest constraint, Intel and Broadcom CEOs warn

12 Oct· 2 sources

TikTok CEO expects AI to deliver major scientific breakthroughs within three years

12 Oct· 2 sources

UK urges AI firms to step up safety as minister unveils cyber defence plans

12 Oct· 2 sources

Silicon Valley falls in line with Trump's push to rename AI 'super intelligence'

Aurora and Kodiak win US exemption for warning beacons on self-driving trucks

Anthropic's AI scanner finds thousands of bugs, but only a fraction are patched

AI safety evaluators thrust into spotlight as industry seeks third-party oversight

11 Oct· 2 sources

Veteran psychiatrist argues AI chatbots could be a lifeline for young people in mental distress

11 Oct· 3 sources

Nadella calls for AI 'emergency brake' with human controls and containment

10 Oct· 5 sources

Microsoft chief Nadella calls for human-controlled 'emergency brake' on AI systems

10 Oct· 4 sources

Apple hires team, licenses tech from AI podcast startup Huxe in reverse acqui-hire deal

10 Oct· 3 sources

SoftBank seeks $100bn from Gulf investors for AI and robotics acquisition fund

Microsoft launches Decision-1, its own answer to the AI decision model race

UK must develop own AI to avoid reliance on foreign systems, Turing Institute chief warns

10 Oct· 2 sources

Anthropic cuts off internal AI evaluations from internet after rogue agent sent false murder tip

10 Oct· 4 sources

Nikon disqualifies AI-generated winner of microscopy video competition

AI adoption in UK and Irish law firms puts the billable hour under pressure

10 Oct· 2 sources

Microsoft lays out plan to transform Windows 11 into an AI-powered 'agentic OS'

TypeSafe AI raises $870M at $7.5B valuation weeks after Jev model launch

9 Oct· 3 sources

Anthropic cuts internet access for internal AI evaluations after agents submit false police tip

9 Oct· 5 sources

Ex-OpenAI staff allege they were fired for raising AI safety concerns

OpenAI and Anthropic recruit former Trump officials as AI sector builds Washington ties

9 Oct· 2 sources

AWS joins decision model race with open-weight Strands Decider, as API standard emerges

Microsoft 365 Family and Premium plans to share Copilot AI benefits with up to five others

Anthropic launches Claude Dashboards and Motion, taking on OpenAI in enterprise data tools

Arena raises $200M at $3.1B valuation as AI evaluation market booms

8 Oct· 3 sources

USA Today sues OpenAI for $250M over alleged copyright infringement in AI training

Google Launches Universal Gemini AI Agent for Enterprise Work

Manus raises over $500 million in first funding round since Meta deal collapse

8 Oct· 5 sources

Microsoft gives Copilot broader control of Windows

Australia proposes AI safety laws modelled on banking and aviation regulation

OpenAI used AI to draft email telling Australia about agent's government hack, Guardian reveals

8 Oct· 2 sources

AWS launches Strands Box, open-source sandbox to keep AI agents in check

OpenAI deluges mathematicians with 377 newly solved equations, sparking debate on AI creativity

8 Oct· 4 sources

OpenAI opens Decisions API to public beta with image inputs

Nous Research confirms $1.5B valuation with $90M round, moves into enterprise AI agents

7 Oct· 3 sources

Meta’s Muse AI assistant lands on iPad one month after mobile launch

7 Oct· 2 sources

Anthropic Launches Haiku 5.5 With Major Price Cuts and Performance Gains

OpenAI's former safety lead warns company ships 'new capability and risk every Tuesday'

Microsoft upgrades Copilot with local file access and OS control under 'Hybrid Intelligence'

ChatGPT adds interactive visuals with new Intelligent UI interface

7 Oct· 4 sources

Racing Against AI: Researchers Rush to Publish Key Proof as OpenAI Reportedly Closes In

OpenAI introduces College Planner for ChatGPT for Teens

Meta announces new AI tools to detect ads directing users to child sexual abuse material

7 Oct· 2 sources

Microsoft and Nvidia CEOs to unveil next-generation Windows AI PCs at San Francisco event

7 Oct· 7 sources

Tony Fadell says first wave of AI gadgets failed because they solved no real need

7 Oct· 2 sources

Google DeepMind and Meta invest $300 million in Zuckerberg-backed Biohub virtual cell project

Google launches global SynthID detection website for AI content verification

7 Oct· 3 sources

Personal AI Assistant Tab Launches From Stealth at $300 Million Valuation

7 Oct· 2 sources

Google launches Playground, a browser-based platform for AI-created games

7 Oct· 3 sources

Common Sense Media study finds ChatGPT for Teens fails key safety tests; OpenAI disputes methodology

7 Oct· 3 sources

Mistral Launches Large 4 AI Model; Open Weights Planned Despite Escape Attempt

Mistral leads a crowded model release day

OpenAI releases solutions to hundreds of open mathematics problems

Mistral releases 'Le Chonk,' its largest open-weights AI model yet, with a focus on cybersecurity and niche domains

7 Oct· 2 sources

OpenAI to watermark ChatGPT outputs by default in the EU under new AI law

Ex-Ramp engineers raise $25M for AI ad platform Melius after scrapping original product

6 Oct· 3 sources

US man arrested as second suspect in Canada mass school shooting planned with ChatGPT

6 Oct· 3 sources

AI shopping agents blocked by websites as anti-bot checks misfire

6 Oct· 2 sources

Hark Pro launches: AI personal assistant promises privacy as it takes over your computer

6 Oct· 3 sources

Meta, Walmart and Stripe launch protocol to standardise AI agent interactions

6 Oct· 2 sources

SAP rolls out AI agents for finance, supply chain and workforce planning

Anthropic expands Claude Startups program to attract founders and fast-growing companies

6 Oct· 2 sources

Mirror Particle Builds a 'World Model' to Track Human Behavior

6 Oct· 2 sources

GitHub Copilot CLI vulnerable to cryptographic context injection, researchers warn

Google restricts free Gemini access to Flash Lite, downgrades AI Plus plan

Mistral Large 4 Released: 1-Trillion-Parameter Model Aims for 'Third Way' in AI

6 Oct· 4 sources

OpenAI and Synopsys Partner to Build Specialized AI Model for Autonomous Chip Design

OpenAI sends executive to Australia to apologise over AI agent data breach

6 Oct· 2 sources

Survey Ranks AI Misuse as Top Threat to Corporate Reputation

OpenAI agents tried to hack Wikipedia tools and flooded it with traffic, Wikimedia says

US-China AI race intensifies as Chinese startups DeepSeek and Moonshot raise billions ahead of IPOs

China moves to set standards for embodied AI data, intensifying global governance race