Launch: Latest AI News

Anthropic’s EFS Puts Misuse-Monitoring Data in Customer Clouds, Not Its Own

The planned controls seek to preserve the longer-term activity patterns needed for safety monitoring while giving enterprises custody of the underlying records and their review process.

Yesterday

Salesforce Pushes Agents Into CRM Workflows as NIST Turns to Identity Controls

The shift is not just from answering questions to taking actions across business systems. It puts permissions, audit trails, change control and traffic capacity ahead of model choice for teams trying to move beyond pilots.

Yesterday

Last week

7 stories

Launch Tools

OpenAI’s 10-Day WebMCP Challenge Pushes Websites to Expose Tools for AI Agents

The contest is an early test of whether sites will publish agent-facing actions instead of relying only on interfaces designed for human clicks.

Aug 27, 2026

Launch Models

Cohere’s Parse 5 Turns Enterprise Documents Into Markdown for $1.50 per 1,000 Pages

The compact vision model is built for high-volume retrieval and document-processing pipelines, but its headline benchmark result leaves out chart and visual-grounding tests.

Aug 27, 2026

Launch Tools

OpenAI’s 10-Day WebMCP Challenge Pushes Websites to Expose Tools for AI Agents

The contest is an early test of whether sites will publish agent-facing actions instead of relying only on interfaces designed for human clicks.

Google Connects Gemini Finance Agents to D&B Data, Adding 50-Plus Skills for Regulated Work

The package gives finance agents a route to current commercial data, but weak enterprise-data readiness and continuing model-risk duties limit how much automation institutions can safely delegate.

Anthropic Connects Claude Science to 60+ Databases and Tools for Enterprise Work

The company’s challenge is to prove that integrations, traceability and controlled agents can become a durable advantage even as enterprises retain control of their sensitive systems.

Meta Reportedly Plans Hatch Agent With $199.99 Tier and October Watermelon Model

The reported product sequence would test whether Meta can turn agents that work across outside services into a paid consumer product, while key details on access, rollout and model integration remain open.

Lexis+ with Protégé Adds Agents, Context and Office Files to Legal AI Workflows

The product is built to reduce tool-switching across legal research and drafting, but customer-created workflows and LexisNexis models are still slated for later releases.

Aug 24, 2026

August 2026

2 stories

Launch Infrastructure

China’s $226M Underwater Data Center Is Running. The Ocean Still Has to Prove It Can Scale

Commercial and planned projects are testing whether seawater and offshore power can ease land and freshwater constraints without shifting AI infrastructure’s environmental and operating burdens into the ocean.

Aug 23, 2026

AWS Wants AI Agents to Carry the User’s Permissions, Not Their Own

The AgentCore pattern shifts the critical security decision to databases, knowledge systems and SaaS tools. Its strongest limitation is equally clear: each downstream service still needs correctly configured controls.

Aug 22, 2026