Since the current bull market began in October 2022, big tech giants have led the rally in U.S. equities. While Nvidia commands the highest market capitalization, it is Google parent Alphabet (GOOGL) that has recently delivered more eye-catching performance. This tech titan, renowned for its global search engine and rapidly growing cloud services platform, has spent the past decade on a massive and steady investment program—one that has now officially come to an end as the company pivots fully toward artificial intelligence.
Looking back over the ten-year period from early 2016 through the end of 2025, Alphabet spent a cumulative total of approximately $346 billion on repurchasing its own stock. From 2018 onward, buyback amounts rose markedly, largely thanks to the U.S. Tax Cuts and Jobs Act of 2017, which reduced the corporate income tax rate from 35% to 21%, freeing up more internal capital for share repurchases. Stock buybacks are generally viewed as a positive signal for companies with stable or growing net income, as they boost earnings per share and enhance appeal to value-oriented investors.
However, this decade-long massive repurchase program was officially terminated on June 1 of this year. Alphabet announced an equity issuance of $80 billion (later increased to $84.75 billion), with $10 billion of that amount placed via private placement to Berkshire Hathaway. The entirety of the raised capital will be deployed toward expanding its artificial intelligence infrastructure, marking a significant shift in the company’s capital allocation strategy.
Although Alphabet continues to maintain a robust advertising business through Google and YouTube, its staggering capital expenditures in AI are now offsetting the scale of its past buybacks. Drawing from historical experience, an all-in bet on AI may yield a mixed set of outcomes.
On the one hand, the long-term prospects are promising. Since the integration of generative AI and large language model solutions into Google Cloud, sales in this high-margin segment have accelerated notably. In the quarter ended March of this year, Google Cloud revenue surged 63% year-over-year, with annualized run-rate sales exceeding $80 billion. Analysts project strong future growth, and the cloud division’s backlog has also seen explosive expansion. Over time, Google Cloud is expected to surpass advertising as Alphabet’s core cash-flow engine.
On the other hand, the early stages of technological development are often accompanied by bubbles. Since the advent of the internet, every revolutionary technology has experienced bubble bursts and correction periods, and AI is unlikely to be an exception. Even if early applications are exciting, it will still take time for companies to optimize AI solutions for improved profitability. Should an AI bubble form and burst, Alphabet may not be immune, but its deep competitive moat and ample cash reserves will make it more resilient than most AI-focused companies.
To support its AI ambitions, Alphabet has also made major moves in the energy sector recently. The company announced a partnership with Cypress Creek Energy to build its largest solar power generation and energy storage project in the United States, located in Arkansas. The project is expected to be developed in phases, and upon full completion, it will feature 2.5 gigawatts of solar capacity and 2.9 gigawatt-hours of energy storage, enough to power more than 315,000 households. The solar modules will be supplied by First Solar (FSLR), with steel procured locally and storage batteries sourced from LG’s U.S. factories. This large-scale project is expected to enter commercial operation in 2029, and Google will purchase all of the initial output at a fixed price.
In addition, Alphabet’s exploration of clean energy extends beyond this project. Earlier this month, the company participated in a funding round for German nuclear fusion startup Proxima Fusion, having previously agreed to purchase electricity from future commercial fusion power plants developed by U.S. fusion firm Commonwealth Fusion Systems.
