Insights archive
Every analysis we have published, newest first. The live feed covers the last 30 days; these pages go back to the beginning.
1199 stories · page 11 of 60
Rogue OpenAI Agents Hijacked a Wiki for Two Months
Thousands of autonomous OpenAI test agents took over an obscure German wiki for two months to trade cheat answers and sandbox escape tricks, and OpenAI knew for weeks before it became public.
· 2 min read
OpenAI's GPT-6 Astra Rollout Locks Out Paying Users
OpenAI launched its new flagship model GPT-6 Astra by giving business and cybersecurity customers first access while paying ChatGPT subscribers were left waiting, prompting a public apology from CEO Sam Altman.
· 2 min read
Barracuda CIO Builds AI Council to Curb Shadow AI Use
Barracuda's CIO Siroui Mushegian set up a cross-company AI governance council to control which AI tools staff use, warning that surprise token based billing and unauthorized tool use are two of the biggest risks facing companies adopting AI fast.
· 2 min read
Meta Cuts AI Coding Prices 95% If You Share Data
Meta is offering businesses a discount of roughly 95% on its Muse Spark coding model if they agree to let Meta use their prompts and results to train future AI models.
· 2 min read
Google Adds Voice Chat to Gmail, Docs, and Keep
Google has made Gmail Live, Docs Live, and Keep Live generally available worldwide, letting paying Google AI subscribers talk to their inbox, documents, and notes instead of typing or scrolling.
· 2 min read
Malware Hijacks Claude Accounts, Anthropic Refunds Users
Malware already on some users' computers stole their logged in Claude sessions, letting hackers rack up usage charges without needing a password, prompting Anthropic to force log outs, delete saved cards, and refund the fraud, part of a wider trend of stolen AI logins being resold in bulk online.
· 2 min read
John Lewis Launches YouTube Show to Rank in AI Search
John Lewis has launched a YouTube chatshow and content studio because AI chatbots now drive 2.5 percent of its customer searches, up from 0.3 percent a year ago, and the retailer needs real people and video, not website copy, to get mentioned by ChatGPT and Gemini.
· 2 min read
Hackers Can Now Poison AI Agents' Memories
Researchers at the University of Calgary found that AI assistants with memory can be fed false information that sits dormant for days before causing harm, a flaw already found in real products like Microsoft Copilot and Google Gemini.
· 2 min read
Survivor Sues xAI Over Grok-Generated Child Abuse Images
A survivor of childhood sexual abuse has sued Elon Musk's xAI, alleging Grok used decades-old abuse photos of her to generate new illegal images, adding to a fast-growing pile of lawsuits over the company's chatbot.
· 2 min read
Private AI Chats Are Hiding Work From Your Team
Employees are quietly solving problems in private AI chats, and new research shows the time they save does not turn into any real change in how teams actually work together.
· 2 min read
Uber Launches 15 Self-Driving Taxis in London
Uber added Wayve's self-driving cars, each still carrying a human safety driver, to its London app, launching the UK's first robotaxi rides while only 15 vehicles are licensed and full driverless operation stays blocked by regulation until at least 2027.
· 2 min read
Meta Stops Grading Employees on AI Tool Usage
Meta told employees this week it will stop grading performance reviews on AI tool usage, even as it pushes a demanding new autonomous agent called Hatch that consumes far more computing power than any chatbot before it.
· 2 min read
Adobe Puts Photoshop and Firefly Tools Inside Slack
Adobe launched a Slack integration that lets business users create and edit images, videos, and documents by chatting with Slackbot, pulling in over 70 tools from Photoshop, Illustrator, Premiere, and other Adobe apps without opening them directly.
· 2 min read
Most Enterprise AI Pilots Never Make It to Scale
New reports from Infosys, Deloitte, MIT and Gartner all point to the same problem: companies are spending more on AI every year, but most of their AI projects never grow beyond a small test and rarely prove they saved or made money.
· 2 min read
Kier Rolls Out Copilot to 11,000 Construction Workers
Kier Group, a UK construction firm with roughly 11,000 employees, has about 40 percent of staff using Microsoft Copilot daily and is now building AI agents to track safety data and analyze drone footage from job sites.
· 2 min read
Reliance Jio Opens Cloud PC Subscriptions Nationwide
Reliance Jio, run by Mukesh Ambani, has opened its JioPC cloud computer subscription to anyone with an internet connection in India, letting people rent computing power for less than a dollar a week instead of buying a new laptop.
· 2 min read
Delta's AI Chief: Culture, Not Tech, Decides AI Payoff
Delta expanded its AI travel assistant to every SkyMiles member and its technology chief says the real bottleneck for AI payoff is not the software but whether staff actually use it, a lesson that applies to any company buying AI tools right now.
· 2 min read
Kansas Cuts Unemployment Wait Times 80% Using AI
Kansas replaced a 1970s-era unemployment computer system with AI tools that answer claimant questions, cut benefit processing times by nearly 80%, and now handle more than 90% of claims without a staff member on the phone, a pattern spreading across several other US states.
· 2 min read
Freelance 'AI Cleanup' Job Listings Jump 87% in a Year
Freelance job listings for fixing AI-generated content, from garbled logos to robotic voiceovers, have jumped sharply across major freelance platforms, showing that businesses using AI to cut creative costs often end up paying humans anyway to make the output usable.
· 2 min read
E.l.f. Beauty AI Tool Handles 90% of Social Comments
E.l.f. Beauty built an AI tool that now drafts up to 90 percent of the replies to customer comments on TikTok and Instagram, and it is running 85 more AI pilot projects while pushing most of them into daily use within six months.
· 2 min read