Picture of the author

区块财富

06/21 09:40

Signal Chief Says AI Assistants Want The Keys To Your Private Life

Signal President Meredith Whittaker warned that users should not treat AI chatbots as friends, confidants or conscious partners in private conversations.

Key Points:

Whittaker said chatbots such as ChatGPT and Claude are not friends or sentient interlocutors.

She said she uses AI only for limited formatting work, not for thinking or writing.

Her strongest warning focused on AI agents that need access to messages, payments, browsers and calendars.

Signal Warning

Whittaker made the remarks in a Bloomberg interview about policy, privacy and Signal, where she was asked about the privacy risks tied to chatbots such as ChatGPT and Claude. “These are not your friends. These are not conscious beings. These are not sentient interlocutors,” she said.

She said she sometimes uses AI tools “to format a document here and there,” but does not ask them questions. Whittaker said she is “very serious” about her thinking and writing, and does not want that process cut off by a system “averaging what’s already out there.”

Her comments fit Signal’s broader position that private communications should not become input for systems built around broad data access. The warning was not framed as a rejection of every AI use, but as a rejection of treating chatbots as trusted personal counterparts.

Also Read: Bitcoin Reclaims $64K On $39B Rally But Spot Demand Weakens

AI Privacy

Whittaker also challenged a scenario from Microsoft AI CEO Mustafa Suleyman, who predicted that users could let Microsoft Copilot handle all their Christmas shopping this year. She said such a system would need wide access to personal data to work as described.

That access, she said, could include a credit card, browser, Signal, the ability to message siblings, a home address and a calendar. “What you’ve just described is a system with very pervasive access across multiple applications and services,” Whittaker said.

She added that, in Signal’s context, this would amount to “a kind of a backdoor.”

Her point was that agentic AI assistants change the privacy question because they do not only answer prompts, they may require permission to act across services.

The debate follows a broader shift in the AI market from chatbots that respond inside one app toward assistants that can plan, buy, message and schedule. That makes access, not only output quality, the central issue in privacy discussions around consumer AI.

Read Next: Saylor Fires Back As Strategy Bitcoin Reserves Beat Debt By $48B
21Compartir

Todos los comentarios0Lo más recientePopular

avatar
Lo más recientePopular