Mistral launches Robostral Navigate, a hardware-agnostic robotics navigation model trained via simulation that utilizes a single camera and language prompts (Benoit Berthelot/Bloomberg)
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Meta launched Muse Image in Meta AI, Instagram, and WhatsApp, and previewed Muse Video — the first media generation models developed by its Superintelligence Labs unit (Meta AI blog).
- Engadget reported the Muse Image model accepts Instagram accounts as a prompt, while Investing.com noted Meta is integrating Muse Image directly into ad creative tools.
- TechCrunch found users were already pushing back over Meta's use of their photos to train the model.
- Forbes framed Muse Image not as a Midjourney competitor but as a play for advertiser creative budgets — a reading absent from most headline coverage.
- Coverage spanned the New York Times, Reuters, Bloomberg, TechCrunch, CNBC, Financial Times, Wall Street Journal and dozens of other outlets.
- The Decoder raised questions about Meta's use of Instagram photos for training, and Meta Superintelligence Labs leaders including Alexandr Wang promoted the launch across X.
Why it matters: Meta is shipping homegrown image generation through Instagram, WhatsApp, and Meta AI while simultaneously wiring the model into ad creative tools — turning the launch into both a consumer product and an ad-tech play, with the training-data controversy giving users and regulators a fresh flashpoint.
