The daily briefing
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.
The biggest AI story today isn’t about who wins a board seat after an IPO—it’s about what the models can now do in the world. Google previewed Genie 4, a world-model system that turns plain-language descriptions into realistic 3D environments and then lets you reshape them with follow-up prompts. The jump from the predecessor is easy to feel in the details: fluid motion looks steadier, complex scenes render more convincingly, and the simulation layer is getting thicker, including more realistic people and added audio effects like crashing waves. Google also says Genie 4 is being studied for robotics training, where it can generate step-by-step prompts (via Gemini 4 Argon) to guide an experiment from description to executable tasks.
Read the full briefing →