Signal Chief Sounds Alarm: ChatGPT and Claude ‘Are Not Your Friends’
As artificial intelligence (AI) becomes increasingly woven into daily life, Signal President Meredith Whittaker is urging users to rethink how they interact with AI chatbots, warning against treating them as trusted companions or confidants.
Whittaker pushed back against the growing tendency to humanize AI systems such as ChatGPT and Claude. Speaking during a Bloomberg interview focused on privacy, policy and the future of technology, she emphasized that users should remain clear-eyed about what these systems actually are.
Asked about the privacy implications of popular AI chatbots, Whittaker said, “These are not your friends. These are not conscious beings. These are not sentient interlocutors.”
Her remarks come as major technology companies race to develop increasingly sophisticated AI assistants capable of carrying out tasks, managing schedules and interacting across multiple digital services.
Concerns Over AI’s Role in Human Thinking
Despite her criticism of overreliance on AI, Whittaker acknowledged that she occasionally uses the technology for limited purposes.
She said she uses AI tools “to format a document here and there.” However, she drew a firm boundary when it comes to relying on chatbots for ideas, analysis or decision-making.
“I don’t ask them questions. I’m very serious about my thinking and writing, and I don’t want the process of working through an idea […] to be foreclosed or eclipsed by the response of a system that’s averaging what’s already out there,” she said.
Whittaker views the emerging vision of agentic AI as more than a technological shift, arguing that it could create a new layer of surveillance by requiring access to vast amounts of personal information.
‘A Kind of a Backdoor’
Whittaker reserved some of her strongest criticism for the idea of AI assistants acting autonomously on behalf of users.
Responding to Microsoft AI CEO Mustafa Suleyman’s prediction that Microsoft Copilot could eventually handle holiday shopping by monitoring conversations, she outlined the extensive permissions such a system would require.
Whittaker said that would mean granting AI access to “my credit card, my browser, my Signal, the ability to message my siblings on my behalf, my home address [and] my calendar”.
“What you’ve just described is a system with very pervasive access across multiple applications and services,” Whittaker said , adding that “in the context of Signal, it would constitute a kind of a backdoor”.
Her comments underscore a widening debate within the technology industry over whether the convenience promised by next-generation AI assistants justifies the level of access they may require to users’ personal digital lives.
