Google Delays Launch of Gemini 3.5 Pro AI Model Due to Coding Performance Issues

Google has postponed the release of its most advanced artificial intelligence model, Gemini 3.5 Pro, by several months after internal tests revealed that its coding capabilities did not meet the company's targets. Originally scheduled for an earlier launch, the development process extended as Google focused on improving the model's performance, particularly in code generation. According to a Bloomberg report, ten current and former Google employees described growing frustration within the company, citing faster progress by competitors OpenAI and Anthropic in AI model development. Internal disagreements among teams and the departure of senior AI researchers have also contributed to challenges.

The delay comes amid intense competition in the AI market, where leading firms strive to enhance model capabilities, reduce operational costs, and expand enterprise adoption. Coding has become a central battleground, making any development setbacks strategically significant. Google updated the model's training data in late June, but the results fell short of internal benchmarks. The company is continuing tests with select partners but has not announced a new launch date for Gemini 3.5 Pro. The news has affected investors, with Alphabet's stock declining amid concerns over Google's ability to keep pace in the AI race.