Our Scans · (FS.1.00) Investment Banking · Weekly Summary

• A 1 MW data center can be used up to 25.5 million liters annually, and according to a Morgan Stanley report, AI data centers are projected to consume 1,068 billion liters by 2028, an 11-fold increase from current levels. Persistence Market Research
• Goldman Sachs expects agentic AI token consumption to grow roughly 24 x by 2030, reinforcing that cost, not capability, is becoming the key constraint. Milk Road
• Morgan Stanley expects China's humanoid sales to more than double in 2026 to around 28,000 units. Republic World
• Morgan Stanley has modeled 8 million working humanoid robots in the US by 2040, with a $357 billion wage impact, and 63 million by 2050 in a much larger long-term scenario. New Market Pitch
• Goldman Sachs expects Brent to remain above US$90 through the end of the year. / USA The Rio Times
• HSBC, Citi, UBS, BNP Paribas, Standard Chartered, Wells Fargo, and eleven others across six continents will test live cross border transactions using tokenized deposits. TechBullion
• Goldman Sachs Research expects behind-the-meter systems to supply a quarter to a third of the incremental data center demand to 2030. / USA RefIndustry
• Morgan Stanley expects $250 billion to $300 billion in debt issuance in 2026 from hyperscalers and their joint ventures, and JPMorgan projects that annual data center securitization could reach $30 billion to $40 billion in both 2026 and 2027. American Enterprise Institute - AEI
• Goldman Sachs now projects approximately $7.6 trillion in cumulative AI infrastructure capital expenditure from 2026 to 2031, covering compute, data centers, and power. American Enterprise Institute - AEI
• Goldman Sachs is even more bullish, forecasting 2030 revenue at approximately $474 billion, with the AI division contributing $322 billion. note
• Goldman Sachs forecasts agentic AI will drive token consumption up 24 x by 2030-reaching 120 trillion tokens per month. TechFlow
• Goldman Sachs expects agent AI to drive a 24 x increase in token consumption by 2030 - to 120 trillion tokens per month. HTX
• JPMorgan Chase & Co. now expects $5.5 trillion of spending tied to AI and data centers through 2030, an increase of $400 billion from its estimate in November. financialpost
• Goldman Sachs projects US data center power demand will reach 41 GW in 2026 and 66 GW in 2027 - more than doubling the 31 GW baseline in 2025. ArentFox Schiff
• Lazard Asset Management explores the outlook for emerging markets, arguing that a synchronized global recovery, vaccine rollout, and supportive policy backdrop could drive stronger performance across equities and debt. The Idea Farm
• Goldman Sachs projects 2030 revenue of $474 billion, with $322 billion from AI. note
• Morgan Stanley projects 2030 revenue of $330 billion, with $190 billion from AI. note

Last updated: 16 July 2026