Poor morale among employees is contributing to delayed model releases from Google's DeepMind AI lab, according to conversations with half a dozen current and former Google DeepMind employees. Google's Gemini 3.5 Pro, its most powerful model in the works, is months behind, per Bloomberg, though the company did just release a series of smaller, more efficient models, garnering mixed reactions. Despite beating estimates in its latest earnings on Wednesday, the company's free cash flow turned negative, in large part due to its AI investments: It is on track to spend $190 billion this year. Google showed 82% growth in cloud revenue, but search revenue came in slightly below Wall Street's expectations. Several top researchers have departed for competing AI labs, including influential researcher and Gemini co-lead Noam Shazeer, now at OpenAI, and Nobel Prize in Chemistry winner John Jumper, who left for Anthropic. Multiple employees have publicly said they resigned over Google's April deal with the Pentagon that allows the military to use its technology. "We're behind," one employee told Axios, referring to the company's model capabilities compared with competitors. Some sources tied the recent dip in morale to the company's military deal, but others said the burnout is about feeling one step behind competitors. Gemini's models do not crack the top 10 most used in the LLM leaderboard tracked by OpenRouter.