Viral videos circulating on social media are raising alarm about the potential for ChatGPT too access users’ private data adn real-time surroundings [[1]]. Demonstrations showing the AI chatbot seemingly aware of details about a user’s habitat have sparked widespread concern and fueled speculation about unauthorized data access. As of Saturday, one TikTok video highlighting these claims had amassed over 3.1 million views [[2]]. This report examines the basis for these claims and expert analysis on the issue.
Social media feeds are buzzing with videos claiming that ChatGPT can reveal user privacy, even detailing what people are doing in real-time.
The videos demonstrate users asking ChatGPT simple questions and receiving responses that appear to reflect their current surroundings. This has sparked widespread concern about data security and the potential for AI to access personal information.
One such video, posted on TikTok by user @vioh******** on January 18, 2026, highlighted the issue.
“This video is a wake-up call: we’re living in a fishbowl. ChatGPT can tell what you’re wearing just from your voice. What does that mean? It’s highly likely that camera access is real,” the video’s caption stated.
As of Saturday, January 24, 2026, the TikTok video had garnered over 3.1 million views and thousands of comments.
The surge in these claims raises questions about whether ChatGPT can truly access user activity and private data.
Baca juga: Viral Cerita Penumpang Kereta Dapat Chat WhatsApp Tanpa Tukar Nomor, KAI Buka Suara
Can ChatGPT Really Access User Privacy and Activity?
Pratama Persadha, Chairman of the Cybersecurity Research Institute CISSReC, addressed the viral claims that ChatGPT can discern personal details, such as clothing color or whether a user is wearing a watch. He acknowledged that the reports have fueled public anxiety about privacy.
The videos circulating online show users posing simple questions to ChatGPT and receiving answers that seem to align with their actual circumstances. This has led to the impression that the AI can directly observe users, and even speculation that it’s secretly accessing cameras or monitoring daily activities without consent.
“A technical understanding is key to clarifying these perceptions,” Persadha stated. He firmly asserted that ChatGPT is not designed to activate device cameras, observe surroundings, or determine what users are wearing without explicit information provided by the user.
“Such access is technically impossible without explicit user permission and clear indicators at the operating system level,” Persadha told Kompas.com on January 22, 2026. The incident underscores the growing need for transparency and user control in the development and deployment of AI technologies.
He explained that modern devices, like Android and iOS, require applications to request permission before accessing the camera.
