Listen to this Post
Artificial intelligence chatbots have exploded in popularity on smartphones, promising convenience, quick answers, and even companionship. But while these apps are fascinating tools, a pressing question remains: how much of your personal data are they collecting? A recent study by VPN provider Surfshark delved into this very issue, examining the privacy practices of the ten most popular AI chatbots on the iPhone App Store. By analyzing the types of data these apps gather, Surfshark has created a ranking from worst to best in terms of user privacy, offering a crucial guide for anyone concerned about digital safety.
Findings
Surfshark’s analysis relied on the personal data disclosures mandated by Apple, supplemented by a careful review of the chatbots’ privacy policies. The research revealed a sobering reality: all ten AI chatbots collect user data, with an average of 14 types of personal information gathered per app. Common data points include location, contact information, browsing history, search history, and even user-generated content. Location tracking, for example, is included in about 70% of the apps.
At the top of the “worst for privacy” list is Meta AI, which collects a staggering 33 out of 35 possible data types—nearly 95% of all available data points. Unique among the analyzed apps, Meta AI collects financial data and other sensitive categories such as racial or ethnic information, sexual orientation, political opinions, disability status, and biometric identifiers. Google Gemini follows as the second-highest data collector, logging 23 distinct types, including precise location, contacts, search and browsing histories, and extensive personal content.
Other apps, including Copilot and Perplexity, also collect sensitive data, though to a lesser extent. The study underscores that free or paid plans offer little to no reduction in data collection. Many chatbots also use your interactions as training data for their AI models, meaning that your questions and conversations could potentially be tied back to you. In contrast, Apple’s own ChatGPT integration via Siri offers better privacy guarantees, anonymizing user queries and preventing them from being used for AI training.
What Undercode Says: Data Collection Trends in AI Chatbots
Surge in AI Popularity and Privacy Risks
AI chatbots have become a staple of daily digital life, but the rapid adoption has outpaced privacy awareness. Surfshark’s findings highlight a troubling trend: even widely used, highly reputable apps are collecting massive amounts of sensitive data. Users often underestimate the breadth of data being harvested, assuming anonymity that does not exist.
Sensitivity of Collected Data
Meta AI and Google Gemini lead in collecting sensitive information that extends beyond typical app usage. Details like political opinions, biometric data, and financial records indicate that AI companies are increasingly positioning themselves not just as service providers but as data aggregators. This has long-term implications for user profiling, targeted advertising, and potentially more severe privacy risks such as identity theft.
Data Usage Beyond Functionality
Most AI chatbots explicitly use user input to train and improve their models. This means that every interaction contributes to a larger dataset that could indirectly identify users, especially when cross-referenced with other collected data. Even paid subscriptions do not mitigate this, suggesting that privacy is often secondary to AI development goals.
Regulatory and Platform Influence
Apple’s policy requiring transparency in data collection has been critical for Surfshark’s analysis, showing how platform regulations can drive accountability. While transparency is improving, enforcement and meaningful user control over data remain limited. Apple’s Siri integration offers a model where user data is anonymized, highlighting the potential for safer AI interactions.
Consumer Awareness and Behavioral Changes
With AI chatbots becoming embedded in search, writing, and productivity tools, users need education on digital privacy risks. Recommendations include choosing apps with strong privacy policies, relying on services that anonymize data, and limiting sensitive queries. Consumers must also advocate for more stringent privacy standards and explore VPN solutions to reduce exposure.
Market Implications for AI Developers
The privacy-conscious segment of users is growing, and developers ignoring this could face reputational and legal risks. Companies like Meta and Google may need to adapt, either by offering more transparent data practices or creating privacy-focused AI models to retain trust.
Comparative Analysis of Data Practices
Surfshark’s ranking also shows the range of privacy practices. Some chatbots collect minimal data, focusing solely on functionality, while others harvest nearly every available data type. This variability suggests a fragmented market where user privacy often depends on corporate philosophy rather than regulatory compliance.
Long-Term Risks of Data Aggregation
Massive data collection can enable deep profiling, predictive analytics, and behavioral targeting. As AI capabilities grow, the risk of misuse increases, particularly for vulnerable populations. Users should approach AI interactions cautiously and prioritize apps with strict data minimization practices.
Tech Innovation vs. Ethical Responsibility
AI developers face a tension between improving models through data and respecting user privacy. Ethical AI design should balance innovation with robust privacy protection, offering users transparency, control, and consent mechanisms.
Recommendations for Users
Prefer AI integrations with anonymization guarantees (e.g., Siri + ChatGPT)
Limit sensitive information input
Regularly review app privacy disclosures
Use VPNs or privacy tools when interacting with AI chatbots
🔍 Fact Checker Results
✅ Surfshark’s ranking is based on Apple App Store disclosures and privacy policies.
✅ Meta AI collects 33 out of 35 data types; Google Gemini collects 23.
❌ Paid subscriptions do not reduce data collection in most AI chatbots.
📊 Prediction
As AI chatbots become ubiquitous in mobile devices, the pressure on companies to enhance privacy practices will intensify. Consumers are likely to shift toward services offering anonymized interactions, potentially boosting privacy-focused competitors. Regulatory bodies may impose stricter requirements for sensitive data handling, and future AI models could integrate built-in privacy-by-design features, offering users transparency and control over their data. Companies ignoring these trends risk losing user trust and facing legal scrutiny.
If you want, I can also create a visually appealing ranking table of all 10 AI chatbots with their data collection score, making the article more engaging and easier for readers to digest. Do you want me to do that?
🕵️📝✔️Let’s dive deep and fact‑check.
References:
Reported By: 9to5mac.com
Extra Source Hub (Possible Sources for article):
https://www.stackexchange.com
Wikipedia
OpenAi & Undercode AI
Image Source:
Unsplash
Undercode AI DI v2
Bing
🔐JOIN OUR CYBER WORLD [ CVE News • HackMonitor • UndercodeNews ]
📢 Follow UndercodeNews & Stay Tuned:
𝕏 formerly Twitter 🐦 | @ Threads | 🔗 Linkedin | 🦋BlueSky | 🐘Mastodon
