California's income tax collections are being driven by enthusiasm around artificial intelligence, with the state's Legislative Analyst's Office warning late last year that AI-fueled revenue represents the "lone bright spot" in the state's fiscal outlook. A new commentary from the R Street Institute highlights the contradiction: while California depends on AI-driven tax revenue, jobs, and economic growth, state lawmakers are advancing regulations that could make it harder to build the data centers needed to power the industry. The state is home to 32 of the world's top 50 AI companies, and its capital-gains-dependent tax system means their success is crucial to funding the Legislature's expanding social programs.
Two major bills are moving forward with strong support in the state Legislature. Senate Bill 886 requires the California Public Utilities Commission to establish tariffs addressing costs tied to transmission, distribution, generation, and interconnection services for data centers, with lawmakers claiming the expansion of such facilities will cause other customers' rates to rise. Senate Bill 887 would provide permitting streamlining—specifically, exemptions from the California Environmental Quality Act process—for companies that rely totally on clean energy within five years. One company has already proposed plans for eight new data centers in the San Francisco Bay Area alone, with local officials reportedly receptive given that region's dependence on the AI economy.
The California Chamber of Commerce argues that existing regulatory processes already ensure new customers bear the costs they cause, and that SB 886 would force data center customers to prefund long-term contracts of at least 15 years for new zero-carbon energy resources. The bill would create an even more bureaucratic rate structure that, according to the Chamber, "introduces significant challenges for project development and financing" in sectors where demand projections, technology needs, and business models evolve rapidly. Meanwhile, likely next governor Xavier Becerra recently complained that California has "hardly any" AI regulations, vowed to "enforce and strengthen" existing rules, and called for "real guardrails" on AI technology.
The commentary points to research showing that increasing electricity demand from data centers has coincided with lower electricity prices rather than higher ones, with a doubling of data center capacity estimated to reduce residential electric rates by 4 percent. PG&E has said that load growth will create billions of dollars in tax revenue and lower electric bills by 10 percent or more by spreading fixed costs across more energy usage. This matters because California's regulated utility system faces enormous fixed costs, and as more residents exit the grid for rooftop solar through net-energy metering, utilities have warned of a death spiral where rising rates on remaining customers incentivize even more people to go solar. The fundamental questions, the commentary concludes, are whether California wants more data centers, whether it wants to power the industry propping up its budget, and whether it wants the good-paying jobs that come with the AI industry—or whether lawmakers will hobble construction even as the state depends on AI revenue to fund its operations.