Google is reportedly developing a next-generation custom artificial intelligence (AI) chip aimed at making its Gemini AI models significantly faster and more energy efficient.
According to The Information, the chip—internally codenamed Frozen v2—is expected to enter production in 2028 and could be six to 10 times more efficient than Google's current AI chips based on the number of AI-generated tokens produced per unit of power.
The report said the new processor is being designed specifically to accelerate Gemini's AI workloads while lowering electricity consumption and operating costs across Google's AI infrastructure.
Although Google did not confirm the existence of the chip, the company said its engineers are continuously working on new technologies to improve AI performance.
"Our teams are constantly researching and experimenting with new innovations to deliver maximum performance and efficiency for our users and customers," a Google spokesperson said.
The company added that while not every research project reaches commercial deployment, its strategy of developing hardware and software together enables it to optimise AI systems more effectively than relying solely on third-party processors.
The reported development reflects a growing trend among major technology companies seeking to build custom AI chips as demand for computing power continues to surge.
Developing in-house processors allows companies to reduce reliance on external suppliers such as Nvidia, whose graphics processing units (GPUs) currently dominate the AI market, while also lowering long-term infrastructure costs.
Google's chip initiative follows similar moves by competitors. OpenAI recently introduced its first custom inference chip, codenamed Jalapeño, while Anthropic is reportedly exploring a semiconductor partnership with Samsung.
The project also aligns with Alphabet's broader AI investment strategy. Earlier this year, the company announced plans to invest between $180 billion and $190 billion in expanding its AI infrastructure and computing capabilities.
Following reports about the new chip, Alphabet shares gained around 3% in early trading ahead of the company's upcoming earnings announcement, reflecting investor optimism over potential long-term cost savings from more efficient AI hardware.
.jpg&w=256&q=75 "Holiday announced on July 22")
