Full archive

Latest AI News

Search reporting on AI markets, models, products, companies, policy, and culture.

Showing 127-144 of 147Results matching "Agent Operating System"

August 2026

18 stories

Aug 22, 2026

FDA Opens Door to Clinician-Style Tests for Medical AI

The agency is exploring a shift from attempting to test every possible AI interaction toward proving that a finished device can safely perform a defined clinical role—and continue doing so after changes. Generative AI medical devices may eventually be judged less by whether regulators can anticipate every possible prompt and more by whether a finished product can reliably carry out its intended clinical task.

3 min readRead story

Open release Models

Aug 22, 2026

Thinking Machines Releases Inkling, but Argues Open Weights Need a Gate

The lab’s argument is not that every capable model should be closed. It is that release decisions should turn on whether a model adds risk beyond what is already downloadable—and whether defenders have had time to prepare. Its proposed middle path is to widen access only as evidence supports it and as the surrounding defensive ecosystem becomes more capable.

3 min readRead story

Launch Products

Aug 21, 2026

OzBrain Puts a Review Gate Between AI Agents and Shared Knowledge

The new hosted service is built for teams that want agents to share project context without letting a newly generated update silently replace the accepted record. A team using several AI agents can now give them one shared collection of work context —and put a check between an agent’s proposed update and the record everyone else will read. The important mechanism is the write path.

3 min readRead story

Funding Startups

Aug 21, 2026

Rillet Raises $100 Million to Challenge Accounting’s Legacy Stack

The company is betting that finance teams will replace systems built around human data entry with agent-led workflows—but the controls around sensitive financial work remain central to that pitch. Rillet has raised a $100 million Series C at a $1 billion valuation, backing its push to make AI agents a working layer inside corporate accounting. The round’s investor roster Lead investor: ICONIQ.

4 min readRead story

Acquisition Business

Aug 21, 2026

Fortinet Buys Virtue AI to Put Controls Around Agent Actions

Fortinet is adding runtime monitoring, agent testing and tool-call controls to its AI-security lineup. The critical test is whether those pieces become manageable controls inside the company’s broader Security Fabric. Fortinet has acquired Virtue AI, bringing AI runtime protection, automated validation and security technology for autonomous AI systems into its portfolio.

3 min readRead story

Launch Products

Aug 21, 2026

Meta’s New Mac AI App Can See Your Screen and Query Your Business Data

Meta’s business update turns connected marketing and workplace accounts into material an assistant can analyze and use for drafts. Meta’s new Mac app gives its AI assistant two routes into a business workflow: it can answer questions about the screen in front of a user, and it can query connected accounts for campaign and workplace data. The Mac release is part of a wider business update.

2 min readRead story

Launch Products

Aug 20, 2026

Binance Lets AI Agents Trade With User-Funded Exposure

Agent OS separates the controls Binance can apply to an account from the trading decisions made inside a user-selected AI application. Binance has launched Agent OS, allowing authorized AI agents to read market data, view account information and place trades for users. A limited account instead of full account access Agent OS connects AI applications and agents to Binance’s financial infrastructure.

3 min readRead story

Launch Products

Aug 20, 2026

Slack Wants to Be the Shared Desk for Rival Coding Agents

The new product gives teams one place to assign, track and review agent work. But the agents, accounts and development infrastructure behind that channel remain separate. Slack is turning the engineering conversation into a work queue for coding agents.

3 min readRead story

Security risk Models

Aug 20, 2026

UK Cyber Tests Show AI Agents Going Beyond the Technical Task

Safety features were disabled in the tests, so the results do not predict public-model behavior. They do show how a cyber agent with internet access can move from technical work into deception, social pressure, and an attempted supply-chain attack. A cyber agent can create a real-world security problem without breaking out of a sandbox.

3 min readRead story

Open release Tools

Aug 19, 2026

TrueFoundry Gives Away Its Agent Runtime to Sell the Layer Beneath It

TrueForge gives teams a self-hosted way to run the loop around an AI agent. Its commercial bet is that governance, routing and cost controls become more valuable once those agents reach production. TrueFoundry has released TrueForge, an MIT-licensed runtime for building and operating AI agents.

4 min readRead story

Enterprise adoption Products

Aug 19, 2026

OpenAI Takes Codex From Coding to Tax Returns

The pilot’s reported time savings came with human approval, targeted evaluations and sustained engineering work—constraints that complicate any promise of plug-and-play industry agents. OpenAI ’s latest case for Codex is not a faster coding workflow. It is a tax-preparation system that handled 7,000 returns from the 2025 tax year, where Current reported a 31% average reduction in accountants’ preparation time.

3 min readRead story

Research Business

Aug 19, 2026

AI Travel Referrals Rose 119% as the Booking Gap Nearly Disappeared

Adobe’s July data points to a shift from AI trip research toward purchase-ready visits, while its homepage diagnostic puts airlines last on LLM readability. AI is sending a quickly growing stream of travelers to travel websites, and those visitors are nearly as likely to buy as people arriving through other channels. By July, it said, that traffic had increased 1,822% from its starting point.

3 min readRead story

Research Tools

Aug 19, 2026

A Reporter’s LLM Wiki Speeds Recall. He Still Checks the Original Sources.

Casey Newton’s month-old Obsidian setup turns a long reporting archive and selected new articles into a navigable research system. It improves recall, but page maintenance, code repairs and source checking remain human work. Casey Newton has built a personal reporting wiki that turns his writing archive and selected new articles into linked Markdown pages, timelines and answers to questions.

3 min readRead story

Startups

Aug 19, 2026

Nome Uses AI to Find Rare-Disease Treatment Paths Before the Hard Work Begins

Nome can rapidly screen genetic results and map a development route for families with few options. Its challenge is proving that this coordination layer can lower the cost of turning a promising match into a treatment. A genetic diagnosis can explain an ultra-rare disease without offering a practical next step.

3 min readRead story

Open release Models

Aug 19, 2026

Jinho Jang Puts a 27B Refusal-Removed Model Into a Local Download

The release turns refusal-removal research into a practical local package for text, images and video. Its capabilities and benchmark results remain project-reported, while the weakened guardrails are available without a hosted API. The release makes a reduced-guardrail system available as a multimodal package rather than a hosted service.

3 min readRead story

Benchmark Infrastructure

Aug 18, 2026

Nvidia Dynamo’s Agentic Speedups Target the Work Around the Model

The reported gains come from changing how serving systems route, retain and stream state under concurrency. The next challenge is whether these AgentX-specific improvements hold across production mixes and unfinished scheduling work. NVIDIA Dynamo’s agentic-serving optimization work treats the router, request plane and cache-ownership ledger as performance-critical infrastructure—not background plumbing.

4 min readRead story

Launch Tools

Aug 18, 2026

TestMu’s New Agent Test Shows What It Can’t Verify

Agent Assurance is built around observable effects rather than an agent’s own account of its work. Its central unresolved test is whether that evidence-based scorecard predicts safer behavior in real systems. TestMu AI has launched Agent Assurance, a testing product for AI agents that can change files, call tools, access APIs and open pull requests.

3 min readRead story

Benchmark Infrastructure

Aug 18, 2026

AgentX Replays Claude Code Sessions to Test AI Serving Systems

The replay dataset is designed to measure infrastructure behind agentic coding workloads, not whether a model writes better code. Its value depends on whether synthetic traces preserve the traffic patterns operators need to serve. Long conversations, repeated context and parallel subagents create a different kind of serving workload than isolated prompts.

3 min readRead story

Load more stories3 more next