Launch: Latest AI News

Broadcom Gives Its vDefend AI Assistant Power to Push Security Rules

The new split is the important constraint: one assistant can explain firewall conditions, while the other can create and apply Distributed Firewall and IDS/IPS rules against detected attacks. The documentation leaves the approval path for those changes unclear.

Aug 22, 2026Tools

Launch Tools

Every Built an AI Copy Editor From Its Editor in Chief’s 30,000 Edits

The internal tool is meant to extend Kate Lee’s editorial judgment across a growing company, not remove the need for expert judgment on difficult work.

Aug 22, 2026Models

Launch Models

Vertex AI Adds Grok 4.6 Preview With a Lower Listed Cache Rate

Google Cloud gives buyers another route to Grok 4.6, but the model remains a Preview offering and the lower cached-input rate is specific to Vertex AI rather than a universal price cut.

Aug 22, 2026Models

Launch Models

DeepSeek Gives V4-Flash a Separate Vision API With a 384-Token Image Cap

The experimental model gives developers several familiar ways to submit images and a fixed upper bound on image-token use. It leaves the harder questions—visual accuracy, latency and reliability—for users to test.

Aug 22, 2026Products

Adobe Makes Firefly’s AI Audio Tools Generally Available

Adobe is combining soundtrack, voiceover and effects generation in one creative suite. Its claim that generated music is commercially safe is central to the pitch—and remains Adobe’s characterization.

Aug 22, 2026Products

Google Lets Publishers Add a Preferred Sources Button to Their Sites

The embed can turn a site visit into a reader’s stated preference across Google surfaces, but Google’s own click-through figure is a past study—not a promise of referrals.

Aug 21, 2026Products

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.

Aug 21, 2026Products

OpenAI Previews Cross-Session Safety Checks Without Keeping Customer Chats

The select-customer preview tries to catch misuse spread across separate API sessions while keeping the provider from retaining the underlying conversations. Its practical test is whether automated signals can supply enough context for safety decisions.

Aug 21, 2026Products

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.

Apple Music Plans to Put AI Labels Where Listeners Can See Them

Apple Music intends to turn provider-submitted AI metadata into a visible label later this year. The policy defines a disclosure threshold, but leaves the method for enforcing it unresolved.

ChatGPT Can Now Send Apple Messages. Its Mac Permissions Set the Terms.

OpenAI says its Messages integration runs locally and does not build an index of all messages. To use it, people must grant ChatGPT access to on-device history, contacts and Mac automation controls.

Anthropic Opens Claude Academy as Free Training for Its AI Products

The free catalog teaches task selection and human review alongside Claude workflows. Whether those lessons improve real deployments remains unmeasured.

OpenAI’s Mac Activity Log Reaches Europe Before Its Docs Catch Up

Computer History gives ChatGPT and Codex a record of selected desktop activity, but conflicting European availability guidance leaves administrators without a clean operational answer.

Ramp Opens Its AI Router to U.S. Customers, With a One-Year Data Default

The new service gives companies one place to direct requests across eight model providers, but its U.S.-only rollout, future pricing and default retention policy set clear boundaries around the offer.

Mistral Gives Enterprise AI Five Ways to Keep Digging Through Documents

The new retrieval loop is designed for answers buried in long files, tables, and linked records. Mistral’s internal benchmarks show large gains, but private enterprise collections remain the more consequential test.

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.

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.

OpenAI Brings ChatGPT Ads to 31 European Markets, With Buying Still Managed

OpenAI’s largest ChatGPT Ads expansion broadens the platform’s European reach, but advertisers must initially buy through OpenAI and its partners before self-service arrives later this summer.