Sergey Brin crashes Google I/O with bold AGI claim: “We’ll build the first one”
Sergey Brin makes unplanned Google I/O appearance, says company is aiming to build the first artificial general intelligence.
Key Takeaways
- Brin’s Surprise Appearance: Sergey Brin returned to the stage after years, hinting at deeper involvement in Google’s AI direction.
- Clear AGI Ambition: Brin publicly stated Google’s goal to build the first artificial general intelligence.
- Strategic Shift: The announcement signals a more aggressive AGI focus across Google’s AI divisions.
- Gemini 2.5 Isn’t AGI: Despite upgrades, Gemini 2.5 is still far from general intelligence.
- Raising the Stakes: Google’s move intensifies the AGI race with OpenAI, Anthropic, and xAI.
In an unexpected turn at Google I/O 2025, co-founder Sergey Brin appeared on stage and made a bold declaration: Google intends to build the world’s first artificial general intelligence (AGI).
The statement, which came after a demonstration of the latest Gemini 2.5 model, marked one of the most direct acknowledgements by Google leadership of its long-term ambitions in the field of AGI—an area typically discussed with caution by major tech firms.
“I wasn’t on the schedule,” Brin said as he joined CEO Sundar Pichai on stage. “But I felt it was important to share where we’re headed. We believe we’re positioned to lead in developing a system that can reason, learn, and adapt broadly—what many would call AGI.”
A shift in tone from Google
Until now, Google had avoided publicly positioning itself in the AGI race, unlike competitors such as OpenAI and Anthropic, who have made AGI their central focus. Brin’s remarks suggest a more assertive shift in Google’s internal narrative, aligning the company more explicitly with the goal of building general-purpose intelligence.
Brin, who stepped back from day-to-day operations in 2019, has rarely made public appearances in recent years. His remarks at I/O signal renewed involvement from the company’s original founders as AI development becomes increasingly strategic.
Gemini 2.5 and growing competition
Earlier in the day, Google showcased Gemini 2.5, the latest version of its large language model, capable of multimodal reasoning, code generation, and image understanding. While the model improves on previous iterations, it does not meet the criteria most researchers associate with AGI.
Still, Google appears to be laying the groundwork. Sources indicate deeper integration between Google DeepMind and Google Research, as well as increased investment in AI alignment and interpretability research.
Brin did not offer technical details or timelines regarding Google's AGI efforts, and the company has yet to publish any roadmap toward that goal.