Google Races to Catch Up in AI Coding Tools with Gemini 3.5 Flash

Google's internal struggle to keep up with rivals Anthropic and OpenAI in the AI coding tools market has led to a surge in customer interest, but the company still doesn't have a clear solution for businesses, according to people familiar with the matter [1]. Google's Gemini model's capabilities are scattered across half a dozen different coding products with different branding, indicating how the company's lack of focus and competing internal efforts have hampered success. Even internally, some Google engineers prefer to use Anthropic's Claude Code, and more concerning, the people said, are the engineers who are struggling to adopt AI coding at all.

Google has made some effort to reduce the internal confusion over priorities, with Chief AI Architect Koray Kavukcuoglu working with Google's main engineering team to unite the company's internal artificial intelligence coding tools under Antigravity, a platform released last year [1]. DeepMind is also devoting more resources to AI coding by forming a new team led by research engineer Sebastian Borgeaud. However, Google still has a philosophical clash between AI researchers who want to move as quickly as possible and more traditional senior engineers who have exacting standards for code quality, former employees say.

In a bid to catch up, Google is rolling out its latest version of Gemini, Gemini 3.5 Flash, a lighter-weight addition to its suite that offers cutting-edge capabilities at half the price of comparable frontier models [2]. According to CEO Sundar Pichai, Gemini 3.5 Flash is "remarkably fast," and the company has strengthened the cybersecurity defenses for the model to make it "less likely to generate harmful content and mistakenly refuse to answer safe queries." Google is also announcing Gemini Spark, a new general-purpose AI agent in the Gemini app that can reason across information in connected apps, and Omni, a world model designed to simulate physical environments.

The market is moving too fast for larger companies like Google to think about it and then move, said Deepti Srivastava, a former Google executive who is founder and CEO of AI startup Snow Leopard [1]. "Speed is your only moat." With Anthropic's recently released Mythos model finding thousands of previously unknown vulnerabilities in the world's software infrastructure, Google's AI portfolio now includes Omni, a world model designed to simulate physical environments, predicting what happens next based on a user's actions. The real question is whether Google's efforts will be enough to keep pace with the rapidly evolving AI coding tools market.