Morgan Stanley named Nvidia its top AI stock pick, projecting that hyperscalers and AI cloud providers will spend $1.58 trillion in 2027 — roughly 60% higher than the $980 billion expected this year. The bank estimates nearly $700 billion of that total could flow to Nvidia, and says 91.7% of the company's 2027 revenue is tied to AI infrastructure demand. Nvidia currently trades at about 15 times fiscal 2028 earnings, and Morgan Stanley says the stock does not need multiple expansion to keep performing.
Key Takeaways
- Morgan Stanley projects $1.58 trillion in AI infrastructure spending by 2027, up 60% from $980 billion this year
- Nvidia could capture nearly $700 billion of that spending, with 91.7% of its 2027 revenue tied to AI demand
- Anthropic is the only major AI company likely to IPO in 2026, with a revised valuation starting at $1.5 trillion
- Bill Gates proposes protecting 40% of jobs from AI replacement through a Human Reserved jobs list
- South Africa saw 75% year-over-year growth in AI course enrollments according to Coursera
- Security fixes are failing more often, with 2.3% needing follow-up patches within 90 days in 2026
- Schneider Electric reports 29% of U.S. manufacturers have AI embedded end to end versus 13% globally
- Trend Micro is investing up to 10 billion yen in Japan's cyber resilience initiative
- SandboxAQ expanded Switch AI to help enterprises monitor AI agents for unsafe behavior and data leaks
- Microsoft continues to invest in AI despite investor skepticism about the trade
Morgan Stanley backs Nvidia amid $1.6 trillion AI build-out
Morgan Stanley named Nvidia its favorite AI stock, with 91.7% of the company's 2027 revenue tied to AI infrastructure demand. The bank estimates hyperscalers and AI cloud providers will spend $1.58 trillion in 2027, up about 60% from $980 billion this year, and nearly $700 billion of that could go to Nvidia. Nvidia trades at 15 times fiscal 2028 earnings, and Morgan Stanley says the stock does not need multiple expansion to keep working. The report also warns of risks including debt burdens, rising component costs, power shortages, and political uncertainty.
Nvidia's AI exposure tied to $1.6 trillion global build-out
A Fortune report by Morgan Stanley says 91.7% of Nvidia's 2027 revenue depends on AI infrastructure spending. Hyperscalers have disclosed more than $1.1 trillion in lease commitments that have not yet started, and about $195 billion of 2027 spending is expected to come from data center leases. The report calls Nvidia the best risk-reward opportunity in the sector and says Wall Street has underestimated past compute cycles by 38% on average. It also notes that Nvidia's share of hyperscaler spending could slip from 18% in 2026 to 16% in 2027, though the company expects to gain share.
Chip stocks fall as investors fear AI spending slowdown
The iShares Semiconductor ETF dropped 11.8% last week, marking its longest losing streak since 2006. Companies like Applied Materials, Lam Research, and Micron also fell sharply as investors worried companies might cut AI spending. However, Bank of America analyst Vivek Arya said the chip sector already prices in a crash that he does not expect, with Nvidia, Broadcom, and Micron trading at about 13 times forward earnings. Cantor Fitzgerald analyst C.J. Muse said chip equipment is effectively sold out, with customers giving at least eight quarters of visibility.
Schneider Electric shares U.S. industrial AI data at PACK EXPO
Schneider Electric released new data at PACK EXPO International 2026 in Chicago showing U.S. manufacturers are adopting AI faster than global peers. Among 1,400 executives surveyed, 29% of U.S. CPG manufacturers said AI is embedded end to end in operations, compared with 13% globally. By 2030, 67% of U.S. respondents expect AI to be core to operations, compared with 37% globally. The report also found U.S. manufacturers face higher operational losses, estimated at 20% of potential revenue, due to labor shortages and aging infrastructure.
Schneider Electric highlights AI adoption gap in U.S. manufacturing
Schneider Electric presented research at PACK EXPO International 2026 showing 52% of U.S. CPG manufacturers have deployed AI across multiple sites, compared with 27% globally. Greg Magdanz, Vice President of OEM Sales, said connecting automation, data, and digital technologies helps manufacturers turn early AI adoption into production efficiencies. The findings also showed U.S. manufacturers expect operational losses to rise to 34% of potential revenue by 2030. Schneider Electric emphasized the need for foundational investments in connected data and open automation to fully capture AI's value.
Anthropic may be the only major AI company planning an IPO in 2026
Anthropic is seen as the only major company likely to announce an IPO in 2026, with 75% odds on the Kalshi prediction market. OpenAI CEO Sam Altman has ruled out a public offering for 2026, citing AI safety concerns. The number of US IPOs that priced this year is down 33% compared to 2025. Anthropic's revised IPO valuation range starts at $1.5 trillion, lower than the prior expectation of at least $2 trillion.
AI attackers are finding ways to bypass security patches faster
AI penetration-testing startup Terra Security found that security fixes are failing more often in 2026. The share of GitHub security fixes needing a follow-up patch within 90 days jumped from about 0.2% between 2019 and 2025 to 2.3% this year. Vulnerabilities with broken fixes are about four times more likely to appear on CISA's Known Exploited Vulnerabilities list. Server-side request forgery and prototype-pollution patches are the most likely to fail.
Bill Gates proposes protecting 40% of jobs for humans from AI
Microsoft co-founder Bill Gates proposed a Human Reserved jobs list that AI would not be allowed to do. He says 40% of jobs could be protected in an extreme version of this plan. Gates pointed to education, mental health care, and caregiving as areas where AI can assist but not replace humans. He also wants to change tax rules so employers paying payroll taxes on human workers are treated differently from those buying robots.
SandboxAQ expands Switch AI for enterprise agent oversight
SandboxAQ expanded its Switch AI platform to help enterprises manage AI agents with greater control. Switch Trust lets companies discover, scan, evaluate, and monitor AI agents in production, checking for prompt injection, data leaks, and unsafe behavior. Switch ROI tracks what each agent costs and delivers, with recommendations to lower spending. Mount Sinai West's MSW Lab is already using Switch Trust for clinical AI applications through its G Board framework.
One person turned themselves into an AI chatbot
San Francisco AI start-up Delphi built a personalized AI version of journalist Arthur, called Arthur AI. The chatbot was trained using 2.5 million of Arthur's own words from books, columns, speeches, and podcasts about the science of happiness. The model is updated weekly as new columns and podcasts are published. Delphi creates interactive digital minds, which are personalized AI clones of real people.
Microsoft Defends AI Investment as Other Investors Doubt
Some investors are questioning the AI trade. Microsoft is bucking the trend by continuing to invest in artificial intelligence. The company stands apart from others who are growing cautious about AI spending.
South Africa Sees 75% Rise in AI Course Enrollments
AI course enrollments in South Africa rose 75% year over year according to Coursera's Global Skills Report 2026. Critical thinking course enrollments grew 132% during the same period. The report draws on data from over 300 million learners across Coursera and Udemy across 98 countries. Trending AI themes in South Africa include generative AI, data and analytics, and agentic AI.
Trend Micro Puts Up to 10 Billion Yen Toward Japan Cyber Resilience
Trend Micro announced the Japan Cyber Resilience Initiative with up to 10 billion yen in support. The initiative helps Japanese organizations strengthen cybersecurity and modernize defenses for the AI era. TrendAI research found 607 publicly disclosed cybersecurity incidents in Japan during the first nine months of 2026, exceeding the 594 reported in all of 2025. Eligible organizations will receive up to 10 days of free professional services and access to TrendAI Vision One AI security capabilities through partners.
Sources
- Morgan Stanley’s favorite AI stock is 92% exposed to the $1.6 trillion build-out it’s bullish on
- Morgan Stanley’s favorite AI stock is 92% exposed to the $1.6 trillion buildout it’s bullish on
- Are Companies Reducing AI Spending? Chip Orders Say Otherwise
- Schneider Electric Releases U.S. Industrial AI Data at PACK EXPO International 2026
- Schneider Electric Releases U.S. Industrial AI Data at PACK EXPO International 2026
- Anthropic is AI trade's last big hope for a major IPO this year
- Security patches need patches now that attackers have AI
- Bill Gates Says Government Should Create ‘Human Reserved’ Jobs List That AI Is Not Allowed To Do’ — 40% of Jobs Should Be Protected for People
- SandboxAQ Expands Switch AI to Give Enterprises Greater Choice, Control, and Accountability Across Their Agentic AI Workforce
- I Turned Myself Into an AI Model
- Some Investors Are Questioning the AI Trade. Microsoft Is Bucking the Trend
- Artificial Intelligence (AI) course enrollments in South Africa rise 75%: Coursera’s Global Skills Report 2026
- Trend Micro Announces Up to ¥10 Billion toward Cyber Resilience Initiative to Strengthen Security for Japanese Organizations in the AI Era
Comments
Please log in to post a comment.