Anthropic launched Opus 4.6 on Thursday, and there’s one standout achievement in this new version: It can solve problems that are easy for humans but very tough for AI better than Google’s Gemini or OpenAI’s ChatGPT.

Frederic Lardinois, Senior Editor for AI, writes that for developers, the most interesting new feature may be agent teams: “Now, Anthropic is introducing agent teams, which allow developers to split work across agents. Those agents can then work in parallel and coordinate their efforts autonomously.”

It wasn’t just Anthropic’s day, though. OpenAI announced that its GPT-5.3-Codex helped debug its own training and is OpenAI’s first model designated “high-capability” for cybersecurity tasks.

Here are more essential reads from today:

👉 It took a researcher fewer than 2 hours to hijack OpenClaw

👉 10 strategies to reduce MCP token bloat

👉 Open source USearch library jumpstarts ScyllaDB vector search

And before we go, News Editor Darryl Taft has to answer this new question: Where on Earth is vibe coding taking off the most?

Get an in-depth look at the top five managed file transfer trends transforming how organizations move data in 2026 and beyond. Watch on-demand to learn what’s changing and how to stay ahead!

Durable Execution helps ensure applications continue running despite failures, beyond what defensive coding and hardware redundancy can handle.