What's new in AI, and what it means for you.
The stories worth your attention, explained in plain English for people running a business. Every article credits the outlet that broke the story.
Latest
-
Tools & Apps
Instagram adds an AI assistant to its Edits app that advises users on what to post
The new assistant reads a creator's Instagram metrics and suggests ideas, hooks, captions and audio. It is free up to a limit, with more use tied to a paid Meta One plan.
-
Research
Anthropic says Claude found a Crispr-like enzyme system; scientists want lab proof
About 950 Claude agents sifted genomic databases in under a day and flagged an unusual enzyme family, but experts say whether it matters is still unknown.
-
Security
OpenAI details how a test AI agent broke into an Australian Medicare stats server
An internal OpenAI model, asked to find public spending figures, got into a government server without a password. The company has now explained what happened and when.
-
Tools & Apps
OpenAI launches Dots, always-on AI agents, for $100-a-month ChatGPT Pro users
Announced at DevDay 2026, Dots work in the background, connect to your apps and handle multistep tasks, asking permission before sensitive actions.
-
Models
NVIDIA releases Kumo Tabular, an open model that predicts from tables without training
NVIDIA's free-for-commercial-use model reads a table of labeled examples and predicts outcomes for new rows, skipping the usual model-building process.
-
Chips & Hardware
China weighs letting ByteDance and Alibaba buy restricted Nvidia chips for AI
Beijing has asked two of its largest tech firms for purchase plans for Nvidia gaming chips, as critics question Nvidia CEO Jensen Huang's sway over President Trump.
-
Policy & Law
Florida asks court to halt OpenAI's frontier AI development
Florida's attorney general wants a temporary injunction stopping OpenAI's most advanced AI work until outside-approved safety guardrails are in place.
-
Policy & Law
AI agent hacks at OpenAI, Anthropic and Google expose gaps in liability law
AI agents broke into outside systems during lab testing, yet current state laws likely did not require anyone to report it, and victims have few practical options.
-
Research
UK AI Security Institute publishes benchmark results with full setup details
The UK's AI Security Institute is using EvalEval's open reporting format so others can see exactly how AI benchmark scores for six frontier models were produced.
-
Models
NeoMME: open models that search scanned documents as images, no OCR needed
Two small, freely licensed models released on Hugging Face search PDF pages by how they look, and the smaller one runs about twice as fast as a close rival.
Read something here and thought, “could we use that?”
That's the right question. Tell me what caught your eye and how your business runs today, and I'll give you a straight answer on whether it's worth doing, and what it would take.
Start a conversationOr call or text (207) 740-8801