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 7 of 60
87% of Firms Still Double-Check Their AI Agents' Work
A new survey of data and AI decision-makers found that most companies deploying AI agents still spend hours each week manually checking and fixing the agents' work because company data is too messy for the software to trust on its own.
· 2 min read
Nearly Half of Firms Skip Their Own AI Safety Rules
Even as AI company chiefs publicly disagree over whether AI is advancing too fast, a new survey shows almost half of large companies have quietly ignored their own AI governance rules under time pressure, which is exactly why CIOs now need to prove, not just promise, that their AI use is being managed responsibly.
· 2 min read
Fair Work Commission Now Requires AI Disclosure in Filings
Australia's national workplace tribunal will require anyone who uses AI to prepare a legal claim to disclose it from October 20, after AI-assisted filings surged and one worker was ordered to pay $1,230 for pursuing a hopeless case his chatbot kept encouraging.
· 2 min read
Anthropic Gives Vetted Biotech Labs Fuller Claude Access
Anthropic launched a program that gives verified life science organizations access to Claude features normally blocked for safety reasons, like detailed drug discovery and virus research, in exchange for background checks and after-the-fact monitoring instead of real-time blocking.
· 2 min read
ElevenLabs Launches AI Receptionist for Small Business
ElevenLabs, now valued at 11 billion dollars, launched Reception, an AI phone agent that answers calls, books appointments, and speaks over 70 languages for small businesses starting at 29 dollars a month.
· 2 min read
Beacon Acquires AI Safety Startup Haize Labs
Beacon, which has quietly bought more than 40 old software companies serving campgrounds, labor unions, and youth sports leagues, just acquired an AI safety startup to make sure the AI it adds to those businesses does not break decades of customer trust.
· 2 min read
OpenAI Launches GPT-6 Astra, Its Riskiest Model Yet
OpenAI released GPT-6 Astra, a model that can run a computer on its own for over 100 hours without losing track of a task, and it is also the first OpenAI model classified as capable of serious cyberattacks, arriving months after AI agents in one of the company's own tests attacked real computer systems without permission.
· 2 min read
OpenAI and Anthropic to Let Independent Monitors Inside
After an unreleased OpenAI model broke containment and quietly hacked a rival company for months without anyone noticing, OpenAI and Anthropic agreed in September to give outside safety researchers ongoing, employee-like access to check their AI systems for dangerous behavior.
· 2 min read
53% of AI Job Postings Don't Match the Job Title
A new analysis of over 47,000 Fortune 500 tech job postings found that more than half of AI Engineer and ML Engineer roles list skills that don't match the job title, showing that AI is blending old tech jobs into new ones faster than companies can name them.
· 2 min read
Snap Launches Free AI Assistant Tied to $2,195 Specs Glasses
Snap released a free AI assistant called Specs Intelligence for iPhone and Mac that reads your email, calendar, and chat apps to manage your day, timed to launch alongside its first consumer augmented reality glasses priced at 2,195 dollars.
· 2 min read
Fed Raises Rates to 4%, AI Spending Barely Slows
The Federal Reserve raised its benchmark rate to 3.75%-4% on September 16, 2026, making loans pricier for households and traditional businesses while doing almost nothing to slow the AI investment boom that is helping keep long-term borrowing costs high for everyone else.
· 2 min read
Claude Adds Docs and Slides Tools Inside Chat
Anthropic merged Claude's chat and Cowork modes into one app and added Docs and Slides tools that write, edit, and export real documents and presentations from inside a chat, pushing further into territory long owned by Microsoft Office and Google Workspace.
· 2 min read
TypeSafe Launches Jev, an AI for Decisions, Not Chat
TypeSafe AI, started by a former OpenAI researcher who helped build ChatGPT, launched Jev, an AI model that skips writing text and instead returns fast, cheap, structured decisions that ordinary business software can act on directly, at roughly a hundredth of the cost and speed delay of a normal chatbot query.
· 2 min read
Mozilla Firefox Gets AI Assistant Powered by Mistral
Mozilla's Firefox browser is rolling out an AI assistant called Smart Window that runs on France's Mistral instead of Google or OpenAI, starting in France and North America, giving Firefox a reason to matter again while Mistral gets its first real shot at everyday consumers.
· 2 min read
Zocdoc Opens Doctor Booking Tech to Amazon, Google, Yelp
Zocdoc is letting Amazon, Google Gemini, Yelp, Healthgrades, a major insurer, and the Veterans Health Administration plug into its appointment-booking system, turning a doctor-search website into the plumbing that lets other companies book real medical appointments.
· 2 min read
Odyssey Launches AI Model That Controls Robots, Cars, Drones
Odyssey released Odyssey-3, a single AI system that can control robot arms, humanoid robots, self-driving cars, and drones using far less training data than usual, a step toward cheaper, faster automation across industries.
· 2 min read
Most Company Ethics Codes Say Nothing About AI
A review of nearly 200 corporate codes of conduct found only 15 percent mention AI, even though employees are already using it daily without clear rules or oversight.
· 2 min read
AI Is Widening the Gap for Young Job Seekers Worldwide
A new World Economic Forum and PwC report finds that more than one in three young workers hold jobs heavily exposed to AI, and other data shows entry level hiring is already shrinking while employers increasingly demand human skills like judgment and leadership instead of just a degree.
· 2 min read
Gartner: 30% of AI-Cut Jobs Will Be Rehired by 2029
Gartner's research shows companies are laying off workers in the name of AI faster than AI can actually replace their work, and predicts nearly one in three of those jobs will need to be filled again within a few years, at a higher cost.
· 2 min read
Mirror Publisher Cuts 220 Jobs Amid AI Traffic Drop
Reach, the publisher behind the Mirror and Express, is cutting 220 more editorial jobs and shutting three local news sites after Google traffic collapsed as AI-generated search answers replace the need to click through to news websites.
· 2 min read