OpenAI Dots Push Spotlights Meta Muse Privacy Woes — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI unveiled its new AI agent Dots at DevDay, with CEO Sam Altman saying the company wants to "set a new standard for privacy in frontier AI," and head of app platform Glen Coates arguing OpenAI is "in a different position to Meta" because Meta lacks a product with 1.2 billion users.
- Meta launched Muse earlier with CEO Mark Zuckerberg calling it "built from the ground up for privacy and security," storing user data on what Zuckerberg described as an isolated Linux VM — though Meta itself can still access that data, per the company's own blog post.
- Muse drew reported security problems within weeks of its App Store-topping debut: a security researcher exposed a zero-day vulnerability (since patched), and 404 Media reported last-minute launch issues including one that could have let users access Meta's internal databases.
- Muse also drew data-handling complaints: an Inc. reporter said it uploaded and read his private messages without permission, a YouTuber said it offered his home address to a stranger via Marketplace, and Wired reported it creates "detailed profiles of all your friends and family"; the agent defaults to letting Meta train on user inputs, with an opt-out.
- Dots has so far avoided major privacy scandals, but its availability is restricted to the $100-and-up ChatGPT subscription tiers — meaning far fewer users are actually testing the privacy claims OpenAI is staking its marketing on.
- OpenAI pitched enterprise customers at DevDay with a framework offering "stronger controls" and zero data retention policy options, explicitly framing privacy differentiation as a wedge against Meta's installed base.
Why it matters: Both companies are staking AI agent launches on mutual privacy accusations, but Meta's Muse has already produced documented security flaws — a zero-day exploit, exposed internal-database risk, and silent access to private messages — within weeks of launch, suggesting the privacy-first positioning is more marketing posture than engineering reality at this stage of consumer AI agents.
Ask SkimNews
