“Now you’re just like, ‘Here’s the magic castle. Build it.’ And it gets done.”
OpenAI’s new “head of preparedness,” Dylan Scandinaro, came from an AGI safety role at the company’s chief competitor. “AI is advancing rapidly,” he wrote in a post on X. “The potential benefits are great—and so are the risks of extreme and even irrecoverable harm. There’s a lot of work to do, and not much time to do it!”
[X (formerly Twitter)]
Things got even weirder on Moltbook, the viral Reddit-style platform, over the weekend.
The plugins are designed to allow Cowork to act like a “domain expert” in areas like sales, legal, finance, marketing, data analysis, customer support, product management, biology research, and more, according to a release. The feature is available now in research preview to all paid subscription tiers.
[Anthropic]
The AI hardware company announced its next bet, dubbed “project cyberdeck”: a portable device specifically designed for vibe-coding. It also announced an over-the-air update to its r1 device that makes the product a “plug-and-play computer controller” that’s able to do agentic tasks on a user’s behalf, and also integrates OpenClaw, the open-source agentic tool formerly known as Moltbot and Clawdbot.
[X (formerly Twitter)]
Hume AI’s CEO, Alan Cowen, and some of its leading engineers will join DeepMind as part of a licensing agreement, Wired reported. The news follows a growing trend of big-ticket “acqui-hires” and licensing deals in the AI industry.
