Noam Shazeer, the CEO and co-founder of Character.AI, is moving back to Google after leaving the firm in October 2021 to form the chatbot startup sponsored by a16z. During his former position, Shazeer led the research team that developed the language model LaMDA (Language Model for Dialogue Applications), utilized in conversational artificial intelligence technologies.
Daniel De Freitas, a co-founder of Character.AI, is also joining Google along with a few other staff members from the firm. General counsel at Character.AI Dominic Perella is taking on an acting CEO role at the company. The majority of the team will remain at Character.AI, the company has announced.
To leverage Character.AI’s technology, Google is also entering into a non-exclusive deal with them.
“I can’t wait to join the Google DeepMind team and work at Google again. I am incredibly proud of all that Character.AI has accomplished in the previous three years. In a statement provided to TechCrunch, Shazeer stated, “I am confident that the funds from the non-exclusive Google licensing agreement, along with the incredible Character.AI team, positions Character.AI for continued success in the future.”
Google announced Shazeer’s arrival to the DeepMind research team but did not disclose De Freitas’s or Shazeer’s precise positions.
Google released a statement saying, “We’re especially excited to welcome back Noam, a preeminent machine learning researcher, who is joining Google DeepMind’s research team, along with a small number of his colleagues.” A Google representative stated, “This agreement will provide Character.AI with increased funding to continue growing and to focus on building personalized AI products for users around the world.
More than $150 million in funding has been raised by Character.AI, most of it from a16z.
“A full stack strategy was necessary to achieve our aim of individualized superintelligence when Noam and Daniel founded Character.AI. Character.AI wrote in its blog post on the change, “To power the experiences that make CharacterAI unique, we had to pre-train models, post-train them, and build a product platform with the ability to reach users globally.”
But in the last two years, things have changed; a lot more pre-trained models are now accessible. In light of these modifications, we believe there is a benefit to utilizing third-party LLMs in addition to our own. This allows us to invest even more money in post-training and developing fresh product experiences for our expanding user base.
It is possible that various regulatory agencies, including the Department of Justice (DoJ) in the EU and the Federal Trade Commission (FTC) in the US, will thoroughly examine these reverse acquisitions. The Competition and Markets Authority (CMA) of the United Kingdom (UK) announced last month that it is investigating Microsoft’s hire of key personnel from Inflection AI to determine whether the tech giant is attempting to evade regulatory supervision. In June, the FTC launched a comparable probe to examine Microsoft’s $650 million acquisition.
