1、Crypto-to-DC Conversion AnalysisMORGAN STANLEY RESEARCHMorgan Stanley&Co.Stephen ByrdGlobal Head of Thematic&Sustainability ResearchStephen.B+1 212 761-3865Jamil OhayiaGlobal Thematic StrategistJamil.O+1 212 761-4776Morgan Stanley does and seeks to do business with companies covered in Morgan Stanle

2、y Research.As a result,investors should be aware that the firm may have a conflict of interest that could affect the objectivity of Morgan Stanley Research.Investors should consider Morgan Stanley Research as only a single factor in making their investment decision.For analyst certification and othe

3、r important disclosures,refer to the Disclosure Section,located at the end of this report.July 15,2025July 15,2025 03:21 PM GMT 2July 15,2025We are Bullish Regarding the Non-Linear Rate of AI Capability ImprovementAn Example:Agentic AI From Model Evaluation&Threat Research,an independent AI assessme

4、nt organization:“We propose measuring AI performance in terms of the length of tasks AI agents can complete.We show that this metric has been consistently exponentially increasing over the past 6 years,with a doubling time of around 7 months.Extrapolating this trend predicts that,in under five years

5、,we will see AI agents that can independently complete a large fraction of software tasks that currently take humans days or weeks.”Source:Morgan Stanley Research 3July 15,2025AI Infrastructure:Total Cumulative Spend of$3 Trillion Through 2028Source:Eikon,Morgan Stanley Research.Past performance is

6、no guarantee of future results.Results shown do not include transaction costs.$2.6 trillion for data centers,including chips and servers,through 2028 110 gigawatts(GW)of power needed through 2028;at$2-3/watt,power plants would cost$210-330b Power grid spend on AI is more difficult to isolate,but is