A major AI player is asking users to share their voice conversations to improve its models, raising questions about data ownership and usage. Anthropic, the company behind the popular AI assistant Claude, has begun prompting users to voluntarily submit their voice recordings for training purposes.
The new request appears when using Claude’s voice features, with a clear prompt asking users to allow Anthropic to use their voice data for AI model improvement. This is entirely optional, and users can choose not to share their voice conversations or opt out at any time in the settings menu. Interestingly, this is separate from Anthropic’s existing option to train on regular chats and coding sessions.
The way it works is that Claude collects audio recordings and voice chat data when users engage with its voice features. This data is then used to improve how Anthropic’s AI models understand and respond to speech. The company claims that users can delete the voice data from settings if they change their mind, but some security experts are already sounding the alarm about the potential implications of sharing sensitive audio recordings.
One expert warns that this could set a worrying precedent for other companies in the AI industry. “If Anthropic is asking users to share their voice conversations without explicit consent, what’s to stop others from following suit?” they say. “We need clearer guidelines and regulations around data ownership and usage in AI development.”
Another concern is the potential for misused or compromised voice recordings. With more companies collecting and storing sensitive audio data, there’s a growing risk of security breaches and unauthorized use. “As we move further into an AI-powered world, we need to be mindful of our digital footprints and consider the long-term implications of sharing personal data,” says a cybersecurity expert.
For users who are concerned about their data, it’s essential to understand what they’re agreeing to when prompted by Claude. By carefully reviewing the settings menu and opting out if desired, users can maintain control over their voice recordings and ensure that their data isn’t being used without consent.
Ultimately, this development highlights the need for greater transparency and accountability in AI model training practices. As companies like Anthropic continue to push the boundaries of what’s possible with AI, it’s crucial that we prioritize user rights and data protection.
Source: Bleeping Computer — 2026-10-04