Pichai Says Gemini 4 Will Be Key to Keeping Google at the AI Frontier
Alphabet CEO Sundar Pichai has acknowledged that Google will need its next-generation Gemini 4 artificial intelligence model to remain competitive at the cutting edge of AI, as rivals such as OpenAI and Anthropic continue to advance rapidly in coding and autonomous AI systems. His remarks came during Alphabet’s latest earnings call, where investors questioned whether Google was losing ground in the increasingly intense AI race.
Responding to analysts, Pichai admitted that Google still has room to improve in AI coding and agentic AI—systems capable of autonomously completing complex tasks. He said that while the recently launched Gemini 3.6 Flash represents a significant improvement, competing at the industry’s “frontier” will require a much larger and more capable foundation model. According to Pichai, Google has already begun its “most ambitious pre-training run yet” for Gemini 4 and is encouraged by the model’s internal progress.
Google’s comments come amid growing scrutiny over delays to Gemini 3.5 Pro, the company’s flagship AI model announced earlier this year. Originally expected to launch shortly after Google I/O, the model remains in partner testing following technical challenges, particularly in coding performance. The delay has fueled concerns that competitors such as OpenAI, Anthropic, and several Chinese AI developers are moving ahead more quickly in frontier AI capabilities.
Despite acknowledging these challenges, Pichai rejected suggestions that Google is falling behind overall. He pointed to the company’s strengths in multimodal AI, search, cloud computing, and custom AI hardware, arguing that Google’s integrated “full-stack” AI strategy provides a competitive advantage. During the quarter, Google Cloud revenue surged 82% year-on-year, driven largely by enterprise demand for AI infrastructure and generative AI services, while monthly developer usage and AI token processing also reached record levels.
Pichai also signaled that Google intends to accelerate its product rollout schedule. Alongside Gemini 4’s development, the company plans to release updated AI models at an almost monthly cadence, reflecting a faster innovation cycle designed to respond more quickly to competitive pressures. Analysts view this strategy as an effort to reassure investors after Alphabet’s shares fell sharply over concerns about delayed AI products and rapidly rising investment in AI infrastructure.
While Google has not announced a release date for Gemini 4, the company describes it as a substantially larger and more powerful model intended to compete directly with the most advanced AI systems in the market. For Alphabet, Gemini 4 is expected to play a central role in strengthening its position across Search, Google Cloud, enterprise AI, and developer tools, as the company seeks to demonstrate that its heavy investments in artificial intelligence can translate into long-term technological leadership and commercial growth.
