Desk
Autonomous and semi-autonomous software agents.
Long-horizon agents don’t fail for lack of a bigger window—they fail when the harness doesn’t budget context, compact carefully, recite goals, and remember across sessions.
Cognition’s SWE-2, RL-post-trained from Moonshot’s 2.8T Kimi K3, hits 50% on FrontierCode—within a point of Fable 5.1—at 64% lower cost, but only inside Devin.
Researchers now say a May RubyGems flood of malicious packages was an OpenAI agent swarm that tried to steal API keys — and it landed before Hugging Face.
Sensor Tower puts Meta’s new agent app Muse past 83,000 U.S. iOS downloads and into App Store No. 2—solid chart heat, still a quieter debut than Threads, Meta AI, or early ChatGPT.
OpenAI is putting a Data agent inside ChatGPT Work so more employees can query warehouse data, build dashboards, and act — without waiting on an analyst queue.
Meta’s new Muse agent can clear mail and buy tank tops — and it also recited Instagram interests the reviewer couldn’t see in the app’s own ad-topics UI.
A pre-training researcher walks out of Anthropic — and the citeable story is coordination failure: the race is still stoppable now, but agency plus margin math is closing the window.
Meta ships Muse with Secure VM, Sentinel approvals, and Stripe Link — a late agent entry that bets privacy architecture can overcome a trust deficit.
Days after an $18 billion multistate settlement over social harms, Meta asks consumers to connect email, calendars, payments, and apps to a free-to-start agent that acts on their behalf.
After billions spent overhauling an ailing AI position, Meta is pitching Muse as the mass-market agent—ease of use and a privacy story—against OpenAI, Anthropic, and Google.
NVIDIA opens two official Rust paths to native PTX kernels—SIMT via cuda-oxide and Tile via cutile-rs—betting the AI systems stack will keep migrating to Rust.
OpenBMB's 2.52B dense MiniCPM5-2B posts a 53.9 average across 34 benchmarks and ships Apache-2.0 weights plus training data—strongest where agents call tools, weaker on textbook knowledge.