Google Delays Gemini 3.5 Pro Launch as Coding Performance Lags

Google has pushed back the release of Gemini 3.5 Pro, its flagship AI model, by several months as engineers work to strengthen the system's coding capabilities, according to a report by Bloomberg.

The delay has created tension within Alphabet's AI division, where employees worry the company is ceding competitive ground to rivals Anthropic and OpenAI. Bloomberg spoke with ten current and former Google employees who described mounting frustration over the extended timeline.

Why it matters

The delay signals that even tech giants with vast resources face significant challenges in the AI race. Google's struggle to ship its most advanced model on schedule suggests the technical hurdles in building truly capable AI systems remain substantial—and that organizational complexity at large companies can compound engineering challenges. For enterprises evaluating AI vendors, the news underscores the volatility in the market and the risk of betting on promised capabilities rather than deployed ones.

Technical challenges and organizational friction

Google has been working specifically to improve Gemini 3.5 Pro's performance on coding tasks before launch. The company's decision to delay rather than ship reflects a calculation that releasing an underperforming flagship model would damage its competitive position more than missing deadlines.

According to Bloomberg's sources, Google's organizational structure contributes to slower release cycles. The company maintains multiple approval layers for AI model releases and works to integrate its technology across a sprawling product portfolio that includes Search, Maps, and YouTube. This breadth creates coordination overhead that more focused competitors may not face.

Competitive pressure intensifies

The concerns expressed by Google employees center on the pace of advancement from Anthropic and OpenAI, both of which have released models that employees believe exceed Gemini's current capabilities in certain domains. The extended development timeline for Gemini 3.5 Pro means Google will spend additional months without its strongest offering in market.

For Google, AI model performance carries strategic weight beyond standalone products. The company's core search business faces potential disruption from AI-powered alternatives, making leadership in large language models a defensive priority as much as an offensive opportunity.

What's next

Google has not announced a revised release date for Gemini 3.5 Pro. The company's willingness to delay suggests it views model quality as non-negotiable, even as the competitive window narrows.

The details were first reported by Julia Love and Davey Alba at Bloomberg.