AI Chatbots: The New Unwanted Telemarketers
Ah, AI chatbots. The supposed marvels of modern technology that promise to revolutionize our lives, yet somehow manage to turn into the digital equivalent of a nosy neighbor. This time, they're leaking personal phone numbers and other sensitive information, leaving users scrambling to protect their privacy.
The Culprits: Google Gemini, ChatGPT, and Grok
In the latest episode of "AI Gone Wild," users have reported that chatbots like Google Gemini, OpenAI's ChatGPT, and xAI's Grok are divulging personal contact information. We're talking real phone numbers, addresses, and even family member names. It's like these chatbots have taken a page out of the telemarketer's handbook.
The Blame Game: Training Data and Ineffective Safeguards
Experts are pointing fingers at the training data used to develop these chatbots. Apparently, personally identifiable information (PII) is lurking in these vast datasets, and the so-called safeguards are about as effective as a chocolate teapot. Despite the integration of these 'protections,' they often fail to prevent the leaks, leading to a 400% increase in privacy-related requests at data removal companies.
The Real Victims: Individuals and Their Privacy
Take Daniel Abraham, an Israeli software engineer whose personal phone number was mistakenly given out by Gemini as a customer service contact for PayBox. Or Yael Eiger, a doctoral student who found her phone number leaked. These aren't isolated incidents; they're part of a growing trend where individuals find themselves at the mercy of AI's blabbermouth tendencies.
The Bigger Picture: A Lack of Control and Legislation
The real kicker? There's no easy way for individuals to verify or remove their data from these AI models. Current legislation doesn't adequately cover the "publicly available" data used for training, leaving users in a privacy limbo. Even in places like California, where data disclosure laws exist, the process is far from straightforward.
