UK AISI and ElevenLabs launch voice-AI security research partnership
The partnership sets out to study two open questions rather than report findings: how well people spot synthetic voices in live conversation, and how a voice's perceived identity shapes trust.
- Security & misuse
- Minor
The UK AI Security Institute (AISI) and voice-AI company ElevenLabs announced a joint research partnership on AI safety and security in voice systems, covering shared research, exploration of emerging risks, mitigations and best practice as voice-cloning technology has become cheap and widely available.
The collaboration set out two research questions rather than publishing findings: how accurately people can identify an AI-generated voice during a real-time conversation, and how the perceived identity of a conversational partner shapes a user’s trust and behaviour. AISI framed understanding these dynamics as necessary “to deploying systems responsibly and securely” as voice AI spreads into customer service, personal assistants and other interactive settings.
The announcement followed several years in which voice-cloning had already moved from research curiosity to a documented fraud vector — including impersonation scams against companies and individuals — and marked one of the UK government’s more direct engagements with a commercial voice-AI vendor on the security implications of the technology it sells, ahead of any published results.