AMD Shares Surge 20% as AI Demand Drives 57% Data Center Revenue Jump

Advanced Micro Devices delivered a striking illustration of artificial intelligence's accelerating hunger for computing power on Tuesday, reporting first-quarter revenue that handily beat Wall Street expectations and sending its shares up 20 percent in premarket trading. The results underscored a broader boom in data centers, the vast facilities packed with servers that underpin everything from cloud computing to the rapid advancement of AI models. Yet the same forces driving this expansion have triggered an intense backlash from prominent progressive lawmakers who warn that the facilities' voracious appetite for electricity and water threatens local resources and climate goals.

AMD said revenue climbed 38 percent from a year earlier to $10.25 billion, exceeding analysts' forecasts. Sales in its data center segment, which includes chips designed for AI workloads, jumped 57 percent to $5.8 billion. Chief Executive Lisa Su described the data center business as now the "primary driver" of the company's growth and expressed confidence that server demand would accelerate as the company scales production. The upbeat forecast for the current quarter, projecting about $11.2 billion in revenue, further signaled that the AI infrastructure buildout shows no signs of slowing.

These numbers arrive at a moment of growing tension in Washington and in communities across the country. Data centers can consume as much electricity and water as a small town, using the latter primarily for cooling the heat generated by thousands of servers. As AI training and inference require ever-greater computational resources, technology companies are racing to construct more of these facilities. Last year, activists blocked or delayed at least 48 proposed projects, according to industry trackers. In one extreme case, a protester fired 13 bullets at an Indiana state lawmaker's home because of his support for new data centers.

The political pushback has reached Congress. Representative Alexandria Ocasio-Cortez and Senator Bernie Sanders have introduced legislation that would impose a nationwide pause on new data center construction while lawmakers study the facilities' environmental and economic impacts. Their argument centers on resource strain: each large facility can draw power equivalent to tens of thousands of homes and significant amounts of water, potentially driving up costs for residents and competing with other societal needs.

Critics of the pause, including policy analysts at free-market groups, contend that such restrictions would hand an advantage to international competitors, particularly China. Paige Lambermont of the Competitive Enterprise Institute argued that slowing American innovation would mean ceding ground to "the authoritarian Chinese version of AI rather than the United States innovators' version." She and others also point to data suggesting that data centers have not yet produced dramatic electricity price spikes. The Institute for Energy Research found no statistically significant relationship between data center concentration and faster increases in electricity rates. In Virginia, which hosts more data centers than any other state, prices have risen more slowly than in some regions with fewer facilities.

Still, most analysts expect prices to rise as AI demand scales. The long-term trajectory depends heavily on how the United States generates the additional power. Natural gas remains the fastest and cheapest option for many developers, yet environmental advocates worry that locking in fossil fuel infrastructure for decades would undermine climate targets. Renewable energy projects, particularly solar and wind paired with battery storage, have expanded rapidly but often face their own local opposition and lengthy permitting timelines.

The tension reflects deeper questions about how society manages technological leaps. AI's potential benefits, from accelerating scientific discovery to improving energy efficiency in other sectors, are frequently cited by industry leaders. Yet those gains depend on enormous upfront energy investments. Data centers already account for roughly 2 to 3 percent of U.S. electricity consumption, and projections from some analysts suggest that share could double or triple within a decade if current trends continue.

Progressive policymakers like Ocasio-Cortez and Sanders have framed the issue as one of priorities. They argue that unlimited corporate expansion should not come at the expense of local communities bearing the costs in higher utility bills or strained water supplies. At the same time, many climate-focused economists contend that an outright pause risks throwing out the potential for AI to help solve climate problems, such as optimizing power grids or speeding the discovery of new battery chemistries.

The AMD results highlight how quickly the market is moving. Su noted strong and increasing confidence that the company's data center AI revenue could reach tens of billions of dollars next year, far outpacing its earlier long-term targets. That kind of growth implies continued heavy investment in physical infrastructure, much of it in states like Virginia, Texas, and Georgia where land and power access have been relatively favorable.

Local governments have taken varied approaches. Some have offered tax incentives to lure data centers, viewing them as economic development anchors that bring jobs and tax revenue. Others have begun tightening water usage requirements or demanding that operators procure renewable energy to match their consumption. The industry itself has invested in more efficient cooling technologies and chips that deliver more computation per watt, though gains in efficiency have historically been outpaced by rising demand.

What remains unclear is whether Congress will act on the proposed pause or whether market forces and state-level regulation will shape the buildout instead. The outcome will influence not only America's competitive position in AI but also its ability to decarbonize its electricity system at the speed many scientists say is necessary. For now, the data centers continue to rise, the chips keep selling, and the debate over how much growth is sustainable grows louder with each earnings report.

The story captures a central challenge of the AI era: balancing legitimate concerns about resource use against the risk of falling behind in a technology that could reshape the global economy. Lawmakers on both sides will face pressure to resolve that tension without resorting to simplistic solutions that either ignore environmental realities or stifle innovation. The coming months of congressional debate, combined with continued strong earnings from companies like AMD, are likely to test which approach prevails.