Google's Gemini 3.5 Pro AI Development Lagging Behind Rivals, Report Says

Google's highly anticipated Gemini 3.5 Pro AI model is reportedly experiencing significant development delays, causing frustration within the company and raising concerns about falling behind rivals like Anthropic and OpenAI. The flagship model is said to be several months behind its original schedule, with Google still working to meet internal quality standards.

Sources, including current and former Google employees, point to coding generation as a major hurdle for Gemini 3.5 Pro. Despite updates to training data in late June aimed at improving this area, test results have not met internal expectations, further pushing back the development timeline. The initial expectation was for the model's debut at Google's I/O developer conference in May, but it did not materialize, and a formal launch date remains unannounced.

Internal organizational structures are also cited as a contributing factor to the delays. With multiple Google teams, including Cloud, DeepMind, and Android, involved in AI development tools, the new model faces a complex review and decision-making process. Employee accounts suggest that internal competition between these teams may be slowing down progress.

Compounding these issues are insufficient computing resources and talent drain. Engineers report difficulties accessing necessary AI tools due to high demand and long waiting times for computational power. Furthermore, some researchers, reportedly disappointed by Google's pace in the AI competition, have moved to companies like Anthropic, exacerbating Google's talent retention challenges.

The setbacks with Gemini 3.5 Pro come at a critical time as the AI landscape intensifies. The continued delays risk allowing competitors to widen their lead, potentially impacting Google's strategic position in the AI market.