Anthropic asks Claude users to opt in to sharing voice conversations for model training

Anthropic has started asking Claude users to voluntarily share their voice conversations to help train its AI models, BleepingComputer reports. The request appears as a prompt when people use Claude's voice features. "Allow us to use your voice data to improve our AI models," it says, adding that audio recordings and voice chat data help Anthropic's models understand and respond to speech. Users can choose "Allow" or "Not now." There is also a dedicated toggle under Settings > Privacy, described as covering "audio recordings and voice chat data" from approved sessions. The prompt tells users they can turn the option off or delete the data at any time. The voice setting is separate from Anthropic's existing option to use chats and coding sessions for training. A user could allow training on voice conversations while keeping regular chats and Claude Code sessions excluded, or the reverse. In the interface BleepingComputer saw, the voice setting was off by default, so users are not automatically opted in. The change matters for privacy because recordings of a person's voice can carry more personal information than text. Separate, per-modality controls let users decide more precisely what they contribute to training.