ElevenLabs just hit $6.6B, but its CEO says the real money isn't in voice anymore

What It Covers

ElevenLabs CEO Mady Stanashevsky reveals the company hit $200M ARR and targets $300M by year-end, while predicting voice AI models will commoditize within two years, forcing strategic pivots toward agents and creative platforms.

Key Questions Answered

• •Model commoditization strategy: Voice models will commoditize in two years, so ElevenLabs treats them as temporary advantage while building durable application layers—creative platforms and conversational agent deployment tools—combining AI models with product interfaces as the new software-hardware integration model.
• •Agent deployment complexity: Successful voice agent implementation requires pronunciation correction, voice selection, system integration with platforms like Salesforce and Google, plus evaluation monitoring and safeguards—not just raw text-to-speech capability, creating defensible moat through workflow orchestration rather than model quality alone.
• •Voice marketplace economics: ElevenLabs operates a voice marketplace where creators share custom voices across 70 languages and earn revenue when others use them, already paying $10M back to 10,000 voice contributors, with top Spanish voice primarily used by English speakers demonstrating cross-language demand.
• •AI content detection framework: Future content verification requires three layers—device-level human authentication encoding, opted-in watermarked AI content for authorized agents like appointment booking, and default assumption that all other content is AI-generated, shifting burden from detecting fake to proving authentic human origin.