Artificial Intelligence

Running AI agents in sandboxes with Microsoft Execution Containers

Microsoft’s open-source sandbox for running untrusted code on Windows, Linux, and macOS has evolved into a cross-platform, policy-based execution layer for agents.

TechCrunch Disrupt: What’s next for AI and software development

Tune into the future of AI and software engineering at TechCrunch Disrupt 2026 (October 13-15 in San Francisco). Register with promo code INFOWORLD25 for a 25% discount.

Google brings predictive AI to BigQuery without the ML training

TabFM could simplify and make predictive analytics cheaper by removing model training and deployment costs, but enterprises may not see the same savings across every workload.

Seven critical vibe coding mistakes — and how to avoid them

All of the productivity gains of AI coding can be lost without careful attention to basic best practices.

From surveillance to safety: How Kazakhstan’s Carpet CCTV is reshaping security

OpenClaw addresses security concerns with a system-wide rewrite

The release spans runtime, plugins, and secret handling as the project describes an evolution into ‘OpenClaw 2.0.’

Why tech needs a new kind of English major

AI can spit out an infinite number of words, but we need humans to use judgment, make meaning, and tell the story—even in a slide deck.

A look at AWS’s agentic toolkit for cloud migration

Amazon Bedrock AgentCore shows how AI agents could make cloud migration faster, safer, and more useful across heterogeneous enterprise platforms.

4 standards solve agent identity. None solves the harder question: Is it still the agent you approved?

A valid agent identity proves almost nothing if the agent can change its behavior without anyone noticing.

Compose Hot Reload gives agents an MCP server to see their own code run

MCP server in Compose Hot Reload allows agents to trigger reloads, take screenshots, inspect the semantic tree, simulate text input, and read application logs.