According to @KobeissiLetter, Nvidia announced a partnership with the US Department of Energy, seven new US government supercomputers, a 1 billion investment in Nokia, and 500 billion in expected business over the next six quarters, and is set to announce AI partnerships with Samsung and Hyundai, source: @KobeissiLetter on X dated Oct 28, 2025. Key trading datapoints reported are the US government partnership and seven supercomputers, pending AI partnerships with Samsung and Hyundai, the Nokia investment, and the six‑quarter 500 billion expected business; the source did not cite any direct cryptocurrency market impacts, source: @KobeissiLetter on X dated Oct 28, 2025.

According to @KobeissiLetter, the U.S. Department of Energy has formed a $1 billion partnership with AMD to build two supercomputers, highlighting intensified federal investment in AI compute capacity that could be relevant for semiconductor traders (Source: @KobeissiLetter on X, Oct 27, 2025). The post characterizes the development as the U.S. joining the AI arms race, underscoring policy-driven demand for high-performance computing that market participants often monitor for revenue visibility and order-book implications at the named vendor, AMD (Source: @KobeissiLetter on X, Oct 27, 2025). No official DOE or AMD confirmation was included in the post; traders should seek primary releases for contract scope, hardware mix, delivery timelines, and counterparties before positioning, as these details materially influence valuation models and sentiment across AI-exposed equities and related crypto narratives (Source: @KobeissiLetter on X, Oct 27, 2025).

According to @KobeissiLetter, the US Department of Energy projects US data centers will consume 325,000-580,000 GWh per year by 2028, roughly equal to the electricity use of about 40 million homes (source: US Department of Energy via @KobeissiLetter). According to @KobeissiLetter, meeting 580,000 GWh would require about 73 average nuclear plants at 8,000 GWh each, yet the US has added only two reactors since 1995 and each takes over 10 years to build, underscoring a capacity shortfall into 2028 (source: @KobeissiLetter). According to @KobeissiLetter, this makes achieving the needed generation by 2028 unlikely at current build rates, signaling a potential power supply constraint for AI build-outs (source: @KobeissiLetter). For crypto, electricity is the dominant operating cost in Bitcoin mining, directly linking miner profitability to power prices and grid tightness (source: Cambridge Bitcoin Electricity Consumption Index). Therefore, the DOE’s projected data center load growth represents a material electricity-market risk factor for BTC miners’ cost curves and hashrate expansion that traders should monitor (source: US Department of Energy via @KobeissiLetter; Cambridge Bitcoin Electricity Consumption Index).