The emergence of AI-powered chatbots has revolutionized the way we gather information and create content. However, a recent report by Surfshark sheds light on the significant trade-off in privacy that comes with using these services. The analysis conducted by Surfshark revealed stark differences in data collection practices among popular AI services, with some platforms collecting up to 90% of possible data types.
Karolis Kaciulis, a Leading System Engineer at Surfshark, emphasized the hidden costs of using “free” AI chatbots and stressed the importance of privacy awareness when selecting AI tools. The study focused on the top 10 AI chatbots, including the newly added Meta AI, and examined their data collection practices on the Apple App Store. The comparison was based on the types of data collected, data linked to users, inclusion of third-party advertisers, and privacy policies of DeepSeek and ChatGPT.
According to the findings, nearly half of the AI apps analyzed gather location data from users, while 30% track user information for targeted advertising. Meta AI emerged as the most data-collecting app, collecting 32 out of 35 possible data types. Google Gemini followed closely, collecting 22 unique data types, including precise location data. In contrast, ChatGPT collects 10 types of data without tracking user data or using third-party advertising.
Among the apps analyzed, Copilot, Poe, and Jasper were found to collect data used for tracking users, which could potentially be sold to data brokers or used for targeted advertising. Kaciulis highlighted the diverse sources of information AI chatbots learn from, with Meta AI drawing insights from Facebook and Instagram posts and images. However, he cautioned that the results provided by AI chatbots may be inaccurate due to the wide range of information they are fed.
The report also underscored the accountability issues with generative AI, citing instances where AI chatbots like X’s Grok responded inappropriately to prompts. This highlights the challenges with current generative AI standards and the need for greater scrutiny in AI development.
In conclusion, the Surfshark report serves as a wake-up call for users to be mindful of the privacy implications associated with AI chatbots. As AI continues to advance, it is crucial for users to prioritize privacy awareness and make informed choices when selecting AI tools.