Sundar Pichai Defends Google's AI Strategy as Gemini 3.5 Pro Delay Raises Concerns

Google CEO Sundar Pichai has mounted a strong defence of the company's artificial intelligence strategy amid growing concerns that it is losing ground to OpenAI, Anthropic and Chinese rivals in the increasingly competitive AI race.

Speaking during Alphabet's earnings call on Wednesday, Pichai acknowledged that Google has areas where it needs to improve, particularly AI coding and autonomous agent capabilities. However, he pushed back against suggestions that the company was no longer competing at the cutting edge of AI development.

The concerns have intensified following Google's delay of Gemini 3.5 Pro, its flagship AI model that was originally expected to arrive in June. The model was anticipated to strengthen Google's position in coding and agentic tasks, two areas where AI companies are competing aggressively.

"We've had clearly frontier models. There are many attributes on which we are still at the frontier; there are areas where we've acknowledged we need to improve and coding and agentic coding is an example of that," Pichai said in response to a question from JPMorgan analyst Doug Anmuth.

Google Bets on Gemini Flash While 3.5 Pro Remains Delayed

Instead of focusing on the delayed Gemini 3.5 Pro, Pichai highlighted Google's Flash family of AI models, which are designed to be faster and cheaper to operate. Gemini Flash is increasingly being deployed across areas including cybersecurity, customer-service agents, enterprise software and data analytics.

Pichai pointed specifically to Gemini 3.6 Flash, released this week, saying the model improved its performance by more than 10 points on a coding benchmark compared with its predecessor while consuming fewer tokens.

Google has also introduced Gemini 3.5 Flash-Lite and Flash Cyber, a model designed specifically for cybersecurity applications. Gemini 3.5 Pro, meanwhile, remains in testing with partners.

The strategy indicates that Google is attempting to compete across different layers of the AI market rather than relying exclusively on increasingly large and expensive frontier models.

Gemini 4 Could Be Google's Big AI Comeback

Pichai also offered an early glimpse at Google's plans for Gemini 4, describing the next-generation model as a very ambitious effort. "I think people will be pleased" when Gemini 4 arrives, Pichai said.

According to the Google CEO, the company is training a significantly larger model designed to compete at the frontier of AI capabilities. He maintained that Google remains very committed and very confident about competing at the industry's leading edge.

Perhaps more significantly, Pichai indicated that Google's Gemini 4 roadmap could involve releasing models at an almost monthly cadence.

Such an approach could help address concerns about Google's model release pace at a time when OpenAI, Anthropic and several Chinese AI companies are introducing new models and capabilities at an increasingly rapid rate.

Google Faces Growing Pressure in AI Coding Race

The delay of Gemini 3.5 Pro has become particularly significant because of the growing importance of AI-powered coding.

Coding has emerged as one of the most commercially promising applications for generative AI, with increasingly capable models able to write software, identify bugs and autonomously complete multi-step development tasks.

Google now faces the challenge of demonstrating that its broader AI ecosystem can compensate for areas where individual Gemini models may trail competitors. The company has significant advantages beyond its models, including Google Cloud, its custom AI chips and the ability to integrate Gemini across products and services used by billions of people.

Alphabet's latest results offered evidence that its broader AI and cloud investments are translating into growth. Google Cloud revenue surged 82 percent during the quarter, significantly exceeding the average estimate of 64 percent.

For investors, however, the question increasingly extends beyond cloud growth. With Google's capital expenditure rising alongside its AI ambitions, Wall Street wants to see whether the company can regain momentum in frontier reasoning, coding and autonomous AI agents.

Pichai's message is that Google is still firmly in the race. Whether the delayed Gemini 3.5 Pro and the more ambitious Gemini 4 can prove that claim will be the considerably harder part.