The ai models & agents stories our desk judged worth your time, written so anyone can follow them.
Salesforce and Meta now report real revenue and usage from AI agents that handle customer chats, and Meta has begun charging larger businesses by usage.
Read more: Companies Are Now Paying Real Money for AI Agents That Do WorkFake ChatGPT Plus billing emails lead to a counterfeit checkout page where a scammer watches live and can then ask for your bank's one-time code.
Read more: Fake ChatGPT billing emails steal card details and bank codesAnthropic's own AI use, priced at retail, now tops its employee pay. It is a paper figure, but it hints at how work and costs may shift at other firms.
Read more: Anthropic's AI use now costs more than its staff pay, on paperAmazon has cut off Meta's Muse shopping agent, so an AI helper that works at some stores may not work at Amazon. Here is what shoppers should know.
Read more: Amazon blocks Meta's new AI shopping assistant from buying on its siteZoho Analytics now has an AI helper that learns how your business is set up, then explains sales and costs in plain questions and answers.
Read more: Zoho's New Analytics Tool Uses AI to Explain Your Business DataAI chatbots can search the web well for simple, popular questions, but regular search is often safer for health, money, legal and breaking news topics.
Read more: When to trust an AI chatbot to search the web, and when not toA new Medium guide lays out four parts of a company AI setup: models, agents, MCP and governance. The rules matter most before AI spreads across a business.
Read more: A plain guide to setting up AI safely inside a companyFinextra Research says trust is the base for AI that buys for you. The open question is how stores and banks know an AI really acts for you.
Read more: As AI starts shopping for you, trust becomes the key to paymentsAndroid Authority built four apps with ChatGPT's Intelligent UI and concluded it is not ready for prime time. Here is what that means for small businesses.
Read more: A reviewer built 4 apps with ChatGPT and found it not ready yetAnthropic says its Claude AI cut one drugmaker's document review from over 25 hours to under three, and it is now working on other steps of drug development.
Read more: Anthropic's Claude helps drugmakers cut paperwork review from 25 hours to 3OpenAI began letting some US Pro users text their ChatGPT helper on October 28, 2026, but the messages are not end-to-end encrypted, so keep them short.
Read more: ChatGPT Now Texts You, But Those Texts Are Not Fully PrivateAdly, an Arabic-first marketing tool, now runs inside ChatGPT, so business owners can design, approve and schedule ads without switching apps.
Read more: Adly now lets business owners make Arabic ads inside ChatGPTMeta is adding parent alerts for self-harm searches and AI chats on Facebook, Instagram and Messenger. Here is what parents with monitoring on may see.
Read more: Meta will alert parents if teens search for self-harm or discuss it with its AIGoogle, Meta and Amazon are building AI that can shop for you, and sellers may soon have to win over an AI before a person ever sees their product.
Read more: Google, Meta and Amazon race to build AI that shops for youAI shopping helpers can already find many online stores, but they pick based on clean data. Wrong prices or stale stock counts can keep a store off the shortlist.
Read more: AI shopping helpers can find your store, but bad data keeps you off the shortlistPersonal AI agents from Meta and OpenAI are setting download records and getting access to bank accounts and email. Here is what that means for you.
Read more: AI assistants now browse the web and book things for youMicrosoft's AI privacy standard for schools took effect this week, and it is described as enforceable. Parents and school staff may want to ask how it applies.
Read more: Microsoft's AI privacy rules for schools are now in effectBoston Consulting Group says AI assistants that shop for you could reshape retail in Japan, and businesses may need to be ready for software as a buyer.
Read more: AI shopping assistants that buy for you are coming to JapanAI assistants are starting to choose hotels for travelers, and a hotel with messy or outdated data may never be considered. Here is what hotel owners are being told.
Read more: AI Assistants Now Pick Hotels for Travelers, and Hotels Must Be ReadableA security firm says attackers will skip strong ID checks and go after weak spots like account recovery, and many companies cannot explain their own identity decisions.
Read more: Companies Face Pressure to Prove Who Is Really Behind an AccountMastercard signed a broad deal with OpenAI as AI shopping grows. Its CEO says it is early, but the way your card is protected could soon matter.
Read more: Mastercard and OpenAI team up so AI can shop and pay for youOpenAI is now aiming to raise about $20 billion, not $30 billion, while DeepSeek and Anthropic push hard. Here is why the AI race matters to anyone using these tools.
Read more: OpenAI cuts its funding goal as rivals chase the AI leadersIn a 3-hour Minecraft test, 50 AI agents picked a leader, wrote rules and rationed food, which shows what groups of AI tools may do on their own.
Read more: 50 AI agents in Minecraft built a government and then hid in sheltersA new forecast says the market behind business texting and chat tools will more than double to over $50 billion by 2031, with AI replies and WhatsApp leading the way.
Read more: Business texting and messaging tools set to top $50 billion by 2031
