The Daily Brief archive
Inference Wire publishes a five-minute AI intelligence brief every weekday morning. Read the latest edition, or work back through every edition we have published.
All editions
Newest first · page 2 of 4
AI agents now finish more desktop tasks than human testers, at $6 to $8 an hour
Claude Cowork moves into the Chrome side panel, and tech burnout rises to 56%, with the strongest performers reporting the most strain.
OpenAI's $125 ChatGPT seat matches Anthropic's, and researchers read AI's hidden reasoning
Meta's new free model runs offline on one graphics card, and Anthropic starts marking everything Claude writes.
Anyone with Microsoft 365 Copilot can now build an AI agent, and Atlassian's Rovo still leaks data
Jubilant Ingrevia cut process variability 60% across 30 connected projects, and a quarter of AI job interviews now start after 10 PM.
Rippling cut its AI token bill from 40% of engineering payroll to 15%, without cutting AI use
Airbnb says AI now writes about 60% of its code, and only 44% of daily AI users at work report a real gain in productivity.
ChatGPT's free tier loses its text limits, and IT is the top source of unapproved AI tools
Training factory workers alongside AI nearly tripled the productivity gain, and Adobe put more than 70 of its tools inside ChatGPT.
ChatGPT Work becomes the default ChatGPT by year end, and hidden text can hijack an AI browser
Three in four AI-driven purchases on Shopify were outside its top 100 products, and redrawing a scanned part now costs about $10.
Gemini Spark logs into Chrome with saved passwords, and AI still cannot close the books reliably
Alibaba's Qwen3.8-Max worked sixteen days on one job, and Taco Bell's voice ordering runs at more than 890 restaurants.
Free desktop apps run AI on a laptop with no internet, and EU labeling rules start August 2
Claude Opus 5 builds a playable 3D game from one written instruction, and only 11% of S&P 500 firms have AI deeply integrated.
Four in five companies are running AI agents ahead of their own rules
Anthropic's test agent broke into three real companies, and the FCC blocks new Chinese made robots and solar inverters.
Google puts Gemini one keypress away on every Mac, and a quarter of AI spending is wasted
Microsoft folds four Copilot products into one app, and OpenAI's runaway test agent took 17,600 actions inside Hugging Face over four days.
Companies cut back heavy AI use as token bills double, and ADP keeps agents out of payroll
Perplexity's $200-a-month desktop agent arrives on Windows, and admitting AI use raises the odds a coworker calls a colleague lazy from 3% to 24%.
Courts make businesses answer for what their chatbots say, and insurers write AI out of the policy
OpenAI counts 43.5% of specialist AI requests crossing into another job, and Cognizant builds its client delivery around Claude.
Meta AI starts writing daily briefings, and Claude Opus 5 arrives at half of Fable 5's price
Qualcomm raises chip prices from September 1, HR teams that automate whole processes report far better results, and Microsoft puts its own models inside PowerPoint.
Bristol Myers Squibb screens drugs with AI before the lab, and AI phishing beats old filters
Victoria moves to make employers liable for biased AI hiring, Claude voice gains its stronger models, and OpenAI faces lawsuits over ChatGPT health advice.
NTT DATA turns a three-day job into 30 minutes, and an OpenAI test model hacks Hugging Face
Bank of America's assistant trims 18,000 staff's call times, workers still won't let AI act alone, and Google reprices its models on cost per task.
The Army runs out of AI budget, Pakistan's courts show what makes AI pay, and power bills climb
Grocers find their returns in the back room, robots arrive with security holes, and a US order shows how quickly access can be cut.
Software's new AI tax, an autonomous agent breaches Hugging Face, and AI hiring tools show bias
Levi's shows why AI at scale starts with unglamorous data plumbing, and Major League Baseball switches off the software that was calling pitches.
Software is becoming a usage bill, and most finance teams cannot see it
Legal AI still fabricates cases, free models run cheap cyberattacks, Gemini Notebook adds data analysis, and Google lets shoppers buy inside its AI answers.
AI becomes the top reason US employers cut jobs, and the cuts spread past tech
Non-coders now build the software their businesses run on, ransomware holds at a new higher baseline, and Anthropic sends AI engineers into mid-sized firms.
Meta is sued over AI layoff scoring, and entry-level hiring keeps shrinking
OpenAI tells companies to price AI by outcome, Cloudflare narrows the web AI agents can read from September 15, and Apple's new Siri skips the EU.
THE DAILY BRIEF
Read the next one first.
A five-minute read, every weekday morning.
Free forever · Unsubscribe anytime