Google Deepmind WIRED has hired the CEO and several top engineers from Hume AI, a startup working on emotionally intelligent voice interfaces, as part of a new licensing deal.
The financial details of the deal are confidential, but Hume AI says the company will continue to supply its technology to other groundbreaking AI labs.
The deal is the latest sign that AI companies expect voice mode to become an increasingly important interface for interacting with customers – and that understanding a user's emotions and mood based on their voice interactions is critical.
Hume AI expects to bring in $100 million in revenue by 2026 as it works with AI labs to tailor AI models to capable and useful voice assistants, said John Beadle, co-founder and managing partner of AEGIS Ventures, which invested in Hume AI. To date, the company has raised $74 million in funding.
CEO Alan Cowen, who has a PhD in psychology, will join Google DeepMind along with about seven other engineers. Cowen and Hume's other AI recruits will help Google DeepMind integrate voice and emotional intelligence into its latest models, according to sources who spoke on condition of anonymity because they are not authorized to speak publicly about the deal.
Hume AI has invested millions in developing models and tools to hone realistic voice interfaces and detect emotions in users' voices. The company trains its models by having experts annotate emotional signals in real conversations. At Google, Cowen and his colleagues will help the tech giant integrate voice and emotion technology into its boundary models, sources say.
“Voice is going to become a primary interface for AI, and that is absolutely where it is going,” said Andrew Ettinger, a veteran investor and executive who takes over as CEO of Hume AI. Ettinger says the company will release its latest models in the coming months.
Beadle of AEGIS Ventures says AI models that can detect a user's emotions and adapt accordingly will become increasingly valuable, not only for consumer devices but also for customer support. “In terms of intelligence, AI models are pretty good at the moment, but from the dimension of general helpfulness – they understand your emotions and can respond in a way that allows you to achieve whatever goal you have – we think there is a huge amount of room for improvement,” says Beadle.
The Hume AI deal could position Google to compete even more aggressively with OpenAI's ChatGPT, which already has a lifelike voting mode. Google also recently teamed up with Apple as part of a multi-year deal that will see Google Gemini power a new version of Siri.
The Hume AI deal is the latest arrangement to blur the line between a partnership and a conventional takeover. Such arrangements allow big tech companies to tap high-quality talent without the government scrutiny that comes with traditional acquisitions — although the Federal Trade Commission recently said it will scrutinize so-called “aqui-hires.”
In 2024, Google paid DeepMind a reported $3 billion to license technology from Character.ai, a company working on lifelike chatbot companions. In similar deals, Microsoft has hired top talent from Inflection; Amazon recruits the team behind Adept; and Meta capture the CEO of Scale AI.
