US Proposes Tax on AI and Crypto

A draft bill in the US Senate threatens to impose fees on data centers that support blockchain networks and AI models if they surpass federal emissions limits. This initiative, led by Senate Democrats Sheldon Whitehouse and John Fetterman, aims to mitigate environmental impacts from increasing energy demand and protect households from rising energy costs. The proposed legislation, named the Clean Cloud Act, requires the Environmental Protection Agency (EPA) to establish emissions performance standards for data centers and crypto mining facilities with over 100 KW of installed power. These standards will depend on regional grid emissions, with a goal of an 11% annual reduction. The bill also includes fines for exceeding emissions limits, starting at $20 per ton of CO2e, with annual increases based on inflation plus an additional $10.

The demand for electricity from crypto miners and data centers is reportedly outpacing the growth of carbon-free energy, with data centers expected to consume up to 12% of the US's total power demand by 2028. Research suggests that the rapid expansion of data centers could produce around 2.5 billion metric tons of CO2 emissions globally by the decade's end. Matthew Sigel from VanEck believes that the legislation unfairly targets Bitcoin miners for their energy consumption. Additionally, this law may conflict with the previous administration's policies aimed at establishing AI safety standards.

The draft bill emerges as Bitcoin miners, such as Galaxy and CoreScientific, are increasingly shifting toward providing high-performance computing power for AI models. Miners have faced challenges due to falling cryptocurrency prices and are diversifying into AI data-center hosting to boost revenue. However, ongoing trade wars could disrupt their business models. Experts warn that aggressive tariffs and retaliatory policies could hinder key players in blockchain networks, emphasizing that the infrastructure supporting crypto could suffer collateral damage during global uncertainties.