Popular YouTuber Gaurav Taneja, better known as Flying Beast, alleged a few days ago, that ChatGPT on his iPhone was secretly accessing his front camera without permission during a voice conversation. Taneja shared a video of his conversation with ChatGPT with other proofs to make his argument. However, tech influencer Abhishek Bhatnagar refuted this claim with strong reasoning from his end.
Gaurav Taneja claims ChatGPT hacked his iPhone
In the six-minute clip, Taneja demonstrates a voice interaction with ChatGPT, asking if there's a 'black spot' on his hand. The AI responds affirmatively, suggesting it's on his middle finger and possibly an ink stain or dirt. Taneja then points to a yellow/orange privacy indicator dot appearing in his iPhone's status bar during the chat, interpreting it as evidence of unauthorised camera access. He claims the AI 'sees everything' and warns viewers about 'dangerous technology.'
Taneja also explains iOS privacy indicators, noting a green dot appears when both camera and microphone are in use (as during video recording), and states that an orange dot signals camera-only access – which he says occurred with ChatGPT. You can see the full video below:
Abhinav Bhatnagar debunked Gaurav Taneja's claims
However, tech content creator Abhinav Bhatnagar swiftly countered the claim in a response video, clarifying key misconceptions about Apple's privacy features.
According to Apple, an orange dot indicates microphone access only, while a green dot signals camera usage (with or without the microphone). The orange dot in Taneja's video is expected behavior for ChatGPT's voice mode, which requires the microphone for real-time audio conversation but does not access the camera.
Bhatnagar emphasises, "Even if ChatGPT lies, the iPhone won't lie." He notes that a green indicator would have appeared if the camera were truly active. Additionally, the AI's detailed response about the hand was not based on visual data – no image was uploaded, and voice mode lacks vision capabilities. Instead, large language models like ChatGPT are programmed to be agreeable, often confirming or guessing details to maintain helpful conversation flow.
Apple also confirmed this on its support page. The tech giant notes that an orange indicator means the microphone is being used by an app on your iPhone, whereas the green indicator means either the camera only or the camera and microphone are being used by an app on your iPhone.
This episode highlights the need for proper verification before diseminating false information, especially if you have a mass following.
