Science & Tech · AI · 1 day ago
AI companies promise privacy as they launch agents that handle personal data
Meta and OpenAI are competing to attract users to AI agents that can handle personal information and perform tasks.
Meta’s Muse stores user data in an isolated virtual computer, but Meta can still access it.
The company says it plans to add a way to prevent that access cryptographically and verifiably later this year.
A security flaw in Muse that could have let someone take control of it was patched, and other security issues were reported before launch.
Muse also allows Meta to use users’ input to train models by default, though users can opt out.
OpenAI has promoted its new agent, Dots, as a trustworthy and secure alternative and shown controls users can set, such as limits on purchases.
These examples show why privacy promises matter: people may share more with agents if they believe their information is protected, but the companies’ claims still need to be tested.
OpenAI and Meta are competing to persuade users that their AI agents can protect personal data.
OpenAI unveiled its Dots agent in late September and said it aims to set a new privacy standard.
Meta launched its Muse agent as a safer successor to OpenClaw, but reports raised concerns about data access and security flaws.
Meta says Muse stores user data in a secure virtual machine and has safety systems to reduce harm.
Meta can still access data in Muse’s virtual machine, though it plans to introduce a way to prevent that access later this year.
- Who
- OpenAI and Meta, whose AI agents are competing for users.
- What
- The companies are making privacy and security promises as concerns grow about how AI agents handle personal data.
- When
- OpenAI announced Dots in late September. Meta launched Muse a couple of months earlier.
- Where
- Not stated
- Why
- The companies are seeking to persuade users to share more information with their AI agents by promising to protect their data.
OpenAI
Meta
Privacy claims
OpenAI
OpenAI says it is focused on having the “most trustworthy, safe, and secure assistant.
Meta
Meta says Muse was built for privacy and security and uses safety systems to reduce the frequency and impact of mistakes.
Data protection
OpenAI
OpenAI’s product lead said the company is in a different position from Meta because Meta has an AI product with 1.2 billion users.
Meta
Meta says Muse stores user data in a secure virtual machine and plans to add a way to cryptographically prevent its access to that data.
set a new standard for privacy in frontier AI
built from the ground up for privacy and security
most trustworthy, safe, and secure assistant
I think we’re in a different position to Meta in that they don’t have an AI product that has 1.2 billion users
Meta launched Muse as a supposedly safer alternative to OpenClaw.
OpenAI announced Dots and highlighted its privacy and security goals.
Sam Altman demonstrated ways users could set rules for their Dots.
- OpenAI agent
- Dots
- Meta agent
- Muse
- Muse predecessor
- OpenClaw
- Muse daily active users
- 600,000 in the US within weeks
- Muse data storage
- A secure virtual machine
- Meta AI product user figure cited by Glen Coates
- 1.2 billion
Sources
Related news
Nadella says advanced AI needs human-controlled emergency brake
Anthropic AI submits false murder tip to Philadelphia police
Anthropic bars repeated cruelty toward its AI models
OpenAI fires three safety researchers after dispute over sensitive information
Gemini in Chrome can summarize pages and compare information across tabs
Major tech figures adopt Trump’s preferred term for AI
Nikon disqualifies AI-assisted microscope video and names a new winner
Microsoft adds security tools and local AI features for Windows
Acemoglu says AI could triple unemployment within a decade
UK privacy watchdog secures data protection commitments from 10 AI companies
Anthropic cuts internet access for all internal AI evaluations
