Google's AI infrastructure leadership stated that the company needs to double its service capacity every six months to meet growing demand. This means achieving a thousand times greater capacity overall within 4-5 years and is considered a strategic priority within the company.
The editorial office reports that Google Cloud Vice President Amin Vahdat, at a company-wide meeting held on November 6, emphasized in a presentation titled “AI compute demand” that infrastructure is precisely the most critical and, at the same time, the most expensive aspect of competition. He noted that this approach aligns with Alphabet's capital expenditures for this year rising to $91–93 billion.
At the meeting, CEO Sundar Pichai stated that the shortage of computing power was delaying product launches, including that these limitations were the main reason why the Veo video generation service could not be opened to a wider user audience. According to him, although the cloud business achieved $15 billion in revenue and a 34 percent growth, the results could have been higher with additional computing resources.
The sector giants — Microsoft, Amazon, Meta, and Google — spending a total of over $380 billion on infrastructure for 2024 alone is considered an unprecedented scale of capital allocation in the market. Experts state that this dynamic, against the backdrop of increasingly intense competition, further highlights strategic risks and sustainability issues in the technology sector.