ARTIFICE DAILY AI + CYBER SECURITY BRIEFING

Artificial Intelligence

Anthropic asks Claude users for voice data

Anthropic added a Claude voice prompt that asked users whether the company could use voice recordings and voice chat data to improve its AI models, with separate privacy controls for different kinds of training.

BleepingComputer

Anthropic had begun showing a prompt in Claude's voice features asking users whether the company could use their voice data to improve its models. The prompt referred to audio recordings and voice chat data.

The interface offered an Allow or Not now choice, and the same preference was available in Claude's Privacy settings. The article said users could change the setting later and could delete the data from settings if they changed their mind.

Anthropic's voice-data option was separate from its existing training choice for chats and Claude Code sessions. According to the article, this meant users could enable one kind of training while leaving the others off.

In the settings view described by the article, the voice option was off by default. That meant users were not automatically included in voice-data training.

Named in this story

Companies

Anthropic
asked Claude users to let it use voice data for model training

Products and systems

Claude
showed a prompt for voice data and has related privacy settings
Claude Code
has a separate training opt-in for coding sessions

How the source tells it

The piece was mostly neutral and explanatory, with a brief reassuring emphasis on the opt-in controls and default-off setting.

  • reassurance the writer highlighted that the feature was optional and disabled by default