Meta Muse Ranks Second in Data Collection Behind Meta AI, Surfshark Warns

Surfshark examined the types of data listed on the Apple App Store for the 13 chatbots. It said Meta's Muse and Meta AI collect almost twice as many types as their competitors: 31 and 33 out of 35 types, respectively. Meta AI remains the top data collector, Surfshark said, with Muse in second place. 'Meta is unbeatable in data collection,' Surfshark said.

Google Gemini ranks third with 24 data types. Surfshark said Muse, Meta AI and Google Gemini are the only tools in the comparison that collect highly sensitive information including ethnic data, sexual orientation, trade union, political opinion, genetic and biometric data. ChatGPT ranks fourth with 17 data types, a 70 percent increase since last year, and now collects health and fitness data, audio recordings and search history, according to the analysis.

Location data is also collected by Gemini, Meta AI, Muse and Perplexity, while ChatGPT, DeepSeek, Claude and Copilot retrieve approximate location information. Surfshark named Pi, Duck.ai and Mistral Vibe as the least intrusive tools. Pi collects only user ID, product interaction and crash data, it said.

Muse is marketed as a tool capable of understanding users' objectives and pursuing them autonomously. Surfshark said the agentic AI app's large-scale data collection practices represent a serious concern for online privacy. The app can collect highly sensitive information from calendars, emails and bank accounts, then combine it with contextual data gathered from user interactions to draw up an exceptionally accurate profile of a person's digital identity, according to Surfshark.

That information can be merged with other datasets for advertising or training purposes, leaving users with very little control over how their data is stored, sold, potentially leaked or corrupted, Surfshark said. It also cited a security researcher's warning that Muse can be manipulated into sending voice recordings to a hacker's server. Users have reported instances in which the tool allegedly read their personal messages without permission and then lied about it, as well as cases in which severe execution faults exposed sensitive information, according to the analysis.

The findings come as Meta faces hefty fines. Surfshark noted that penalties range from $18 billion related to addiction among minors to €1.2 billion for improper handling of user data in 2023. Meta is also known for smart glasses that can record people without their consent and that have sparked controversy over reported harassment and stalking.

Surfshark argued that fines may not deter large technology companies. It cited Proton research finding that $7.8 billion in privacy and competition fines imposed on Big Tech in 2025 could be collectively paid by these firms in less than a month. Costa argued that regulators are escalating the financial stakes, shifting toward massive penalties like Anthropic's $1.5 billion and Meta's $1.4 billion settlements.