Big Tech companies are pouring massive amounts of money into AI infrastructure, and investors are starting to worry about the financial toll. Alphabet, Microsoft, Amazon, Meta, and Oracle plan to spend nearly $780 billion on AI infrastructure this year alone. That figure is expected to exceed $1 trillion annually from 2027 through 2030. The spending spree has already hit balance sheets hard — Alphabet reported negative free cash flow of $5.9 billion in Q2 and carries $98.2 billion in long-term debt, while Meta's free cash flow dropped 91 percent year over year.
Not everyone is panicking though. AllianceBernstein warns that rising interest rates could slow down AI capital spending significantly in 2027 and 2028, potentially reshaping how investors value AI stocks. On the other hand, Morgan Stanley projects AI capital spending could reach $1.58 trillion by 2027, with AI company earnings growing around 25 percent. The firm believes the recent U.S. stock correction is nearing its end, especially since Micron and Broadcom met its high-quality earnings criteria.
Jim Cramer, a well-known CNBC commentator, urged investors not to rush back into AI stocks just yet. He said four things need to happen first: technical charts improve, Big Tech confirms continued AI spending, the midterm elections pass in November, and Anthropic completes a successful IPO. Until then, Cramer thinks it is too early to get aggressive on AI data center stocks.
Some investors are making bold bets regardless. David Tepper's Appaloosa Management has roughly 31 percent of its portfolio in Micron and Amazon. Micron's revenue jumped 379 percent to $54.2 billion in Q4 fiscal 2026, and its shares are up over 200 percent this year. Amazon reported Q2 revenue of $200.6 billion, up 20 percent. Tepper sees both companies as well positioned to benefit from strong AI demand, particularly as memory chip shortages are expected to last through 2028.
Alphabet itself remains a popular pick for downturns. The company earned $119.8 billion in revenue in Q2 with 24 percent growth, and Google Cloud grew 82 percent to $24.8 billion. It generated $39.1 billion in operating cash flow and maintains a strong balance sheet. Meanwhile, smaller players like Iren are positioning themselves as safer bets, with a 5.8 gigawatt pipeline and a target of $4 billion in annual recurring revenue by year end.
In chip news, Tenstorrent raised $1.5 billion in a Series E round, doubling its valuation to $6.2 billion. CEO Jim Keller said the company chose new funding over a potential sale and demonstrated that its Galaxy servers can handle DeepSeek AI models. The company competes directly with Nvidia and AMD and now has 1,150 employees.
On the software side, Mistral AI launched Timestral, a family of four time series models ranging from 5M to 209M parameters. The 209M model outperforms Google's TimesFM-3 on fev-bench and runs entirely on standard server CPUs without needing a GPU. The 5M model can forecast one day ahead in just 10 milliseconds, while the largest model takes 140 ms.
ExlService Holdings is pushing agentic AI deeper into insurance and healthcare through its new LifePRO Agentic Suite and upgraded EXL Payment Integrity platform. The stock has returned 40.2 percent over 90 days, though some investors question the track record. Tufin also updated its security suite to version 5.4 with expanded AI-ready network security features, including broader support for SDN, NFV, and cloud networks along with new compliance tools.
Real estate is getting in on the action too. Data center users accounted for 14.4 percent of industrial leasing across six major U.S. markets in 2025, a 44 percent increase from the prior year. AI firms occupy over 7 million square feet in San Francisco and accounted for 34 percent of tech office leasing in London. Power access is increasingly the key factor determining property value, according to GlobeSt's analysis of six AI real estate strategies.
On the security front, Dito joined Google Cloud's Managed Security Services Provider program. The company offers 24/7 security operations powered by Google Security Operations and Mandiant escalation, with agentic threat detection and Gemini-enabled alert summarization. Dito guarantees critical threats are acknowledged within 15 minutes and promises a 70 percent reduction in false positives within three weeks.
Key Takeaways
- Five hyperscalers — Alphabet, Microsoft, Amazon, Meta, and Oracle — plan to spend nearly $780 billion on AI infrastructure in 2025, with spending expected to exceed $1 trillion annually from 2027 through 2030.
- Alphabet reported negative free cash flow of $5.9 billion in Q2 and carries $98.2 billion in long-term debt, while Meta's free cash flow fell 91 percent year over year due to heavy AI spending.
- David Tepper's Appaloosa Management has about 31 percent of its portfolio in Micron and Amazon, with Micron's revenue jumping 379 percent to $54.2 billion and shares up over 200 percent this year.
- Tenstorrent raised $1.5 billion in a Series E round, doubling its valuation to $6.2 billion, and demonstrated its Galaxy servers can handle DeepSeek AI models.
- Mistral AI launched Timestral, a family of four time series models (5M to 209M parameters) that run on standard CPUs without GPUs and outperform Google's TimesFM-3 on fev-bench.
- AllianceBernstein warns that rising interest rates could significantly slow AI capital spending in 2027 and 2028, while Morgan Stanley projects AI spending could reach $1.58 trillion by 2027.
- Jim Cramer advises investors not to rush back into AI stocks until technical charts improve, Big Tech confirms spending, midterm elections pass, and Anthropic completes a successful IPO.
- Data center users accounted for 14.4 percent of industrial leasing across six major U.S. markets in 2025, a 44 percent increase from the prior year.
- ExlService Holdings launched its LifePRO Agentic Suite for insurance and healthcare, with shares up 40.2 percent over 90 days.
- Dito became an approved Google Cloud MSSP, offering Gemini-enabled alert summarization and guaranteeing critical threats are acknowledged within 15 minutes.
Alphabet is a strong AI stock pick for downturns
Alphabet is recommended as a top AI stock to buy before a market crash. Its products like Google Search, YouTube, and Google Cloud are used by billions of people and businesses. In Q2, Alphabet earned $119.8 billion in revenue with 24 percent growth. Google Cloud grew 82 percent to $24.8 billion. The company generated $39.1 billion in operating cash flow and has a strong balance sheet.
David Tepper's Appaloosa heavily invested in Micron and Amazon
Billionaire David Tepper's Appaloosa Management has about 31 percent of its portfolio in Micron Technology and Amazon. Micron makes memory chips for data centers and saw revenue jump 379 percent to $54.2 billion in Q4 fiscal 2026. Its shares are up over 200 percent this year. Amazon reported Q2 revenue of $200.6 billion, up 20 percent, and is a leader in cloud computing and AI. Both stocks benefit from strong AI demand.
David Tepper backs Micron and Amazon AI stocks
David Tepper's Appaloosa Management portfolio is about 31 percent invested in Micron Technology and Amazon. Micron posted non-GAAP earnings per share of $33.42, up 1003 percent year over year, and benefits from a memory chip shortage expected to last through 2028. Amazon increased its stake by 15.7 percent in Q2 and has cloud demand visibility through 2028. Both companies are well positioned in the growing AI industry.
Hyperscalers face risk from trillion dollar AI spending
Five major hyperscalers including Alphabet, Microsoft, Amazon, Meta Platforms, and Oracle will spend nearly $780 billion on AI infrastructure this year. Spending is expected to exceed $1 trillion annually from 2027 through 2030. Alphabet carried $98.2 billion in long-term debt as of June 30, while Meta carried $83.7 billion. Alphabet reported negative free cash flow of $5.9 billion in Q2, and Meta's free cash flow fell 91 percent year over year. Investors face high financial risk and uncertainty about returns on this massive spending.
Iren is the AI stock to own through bubble scares
Iren is recommended as the one AI stock to hold through the next market bubble scare. The neocloud provider has a pristine balance sheet and a 5.8 gigawatt pipeline. It expects to deliver 300 megawatts by 2026 and 800 megawatts by end of 2027. Iren is targeting $4 billion in annual recurring revenue by year end. Recent strong results from Micron and Taiwan Semiconductor suggest tech companies are still spending heavily on AI.
Mistral's Timestral family tops time series benchmarks
Mistral released Timestral, a family of four time series models ranging from 5M to 209M parameters. The 209M model ranks first among zero-shot models on GIFT-Eval, TIME and fev-bench. All models run on standard server CPUs without needing a GPU. The 5M model forecasts one day ahead in 10 ms while the 209M takes 140 ms.
Mistral AI launches Timestral forecasting model family
Mistral AI released Timestral, its first time series foundation model family with four sizes from 5M to 209M parameters. The 209M model outperforms Google's TimesFM-3 on fev-bench with a SQL skill score of 50.39 versus 48.99. It uses synthetic-first pre-training and decoupled horizon forecasting. The models use open weights and run on CPUs with 21 quantile outputs.
ExlService Holdings expands agentic AI in insurance and healthcare
ExlService Holdings is expanding agentic AI across its insurance and healthcare offerings. The company launched its LifePRO Agentic Suite and upgraded its EXL Payment Integrity platform. Shares trade at US$39.52 after a 40.2% return over 90 days. Bulls see the AI rollout as justification for a higher valuation while bears question the track record.
AI investment faces critical moment as rates rise
AllianceBernstein says the AI trade is at a critical crossroads because rising interest rates threaten to slow capital spending. The firm expects significant deceleration in capex growth in 2027 and further slowdown in 2028. This could reshape how investors view AI-related stocks and infrastructure investments.
Dito joins Google Cloud MSSP program with AI security
Dito became an approved Google Cloud Managed Security Services Provider. The company offers 24/7 security operations powered by Google Security Operations and Mandiant escalation. Its services include agentic threat detection and Gemini-enabled alert summarization. Dito guarantees critical threats are acknowledged within 15 minutes and promises a 70% reduction in false positives within 3 weeks.
Tenstorrent raises $1.5 billion, doubles valuation to $6.2 billion
Tenstorrent raised $1.5 billion in a Series E funding round, valuing the AI chip startup at $6.2 billion. The company's valuation more than doubled from $2.6 billion in 2024. CEO Jim Keller said the company chose new funding over a potential sale and demonstrated its Galaxy servers can handle DeepSeek AI models. Tenstorrent has 1,150 employees and competes with companies like Nvidia and AMD.
Tufin releases TOS 5.4 with expanded AI-ready network security features
Tufin launched version 5.4 of its Tufin Orchestration Suite, adding features for AI-ready network security. The update includes broader support for SDN, NFV and cloud networks, improved real-time network visibility and new compliance tools for PCI-DSS, HIPAA and GDPR. CEO Reuven Hernovich said the release helps enterprises keep up with rapid digital change. Tufin provides network security automation to large organizations worldwide.
Jim Cramer warns investors not to rush back into AI stocks
CNBC's Jim Cramer advised investors against returning to AI stocks too quickly. He said four things must happen first: technical charts improve, Big Tech confirms continued AI spending, the midterm elections pass in November, and Anthropic completes a successful IPO. Cramer said it is too early to dive back into AI data center stocks. He wants to see clear signs of market recovery before becoming aggressively bullish on the sector.
Chip stocks rally as AI hardware correction may be ending
AI hardware stocks rebounded on October 19, with SanDisk rising over 7%, Intel gaining over 5% and Micron up more than 4.3%. Morgan Stanley strategist Michael Wilson said the U.S. stock correction is nearing its end if corporate quality stays high and earnings estimates keep rising. Morgan Stanley projects AI capital spending will reach $1.58 trillion in 2027, with AI company earnings growing about 25%. Micron and Broadcom met Morgan Stanley's high-quality and upward earnings revision criteria.
Six AI real estate strategies go beyond data centers
GlobeSt outlined six AI real estate plays including industrial conversions, inference clusters, AI offices, power-ready land, battery storage and construction housing. Data center users accounted for 14.4% of industrial leasing across six major U.S. markets in 2025, a 44% increase from the prior year. AI firms occupy over 7 million square feet in San Francisco and accounted for 34% of tech office leasing in London. Power access is increasingly the key factor determining property value. Target Hospitality, Civeo and Corporate Mobile Housing serve construction crew housing needs.
Sources
- If I Could Only Buy 1 Artificial Intelligence (AI) Stock Before the Market Crashes, It Would Be This One
- Billionaire David Tepper’s Appaloosa Management Has 31% of its portfolio Invested in These 2 Artificial Intelligence (AI) Stocks
- Billionaire David Tepper’s Appaloosa Management Has 31% of its portfolio Invested in These 2 Artificial Intelligence (AI) Stocks
- The Trillion-Dollar AI Bet: What It Means for Hyperscaler Investors
- If I Could Only Own 1 Artificial Intelligence (AI) Stock Through the Next Bubble Scare, It Would Be This One
- Mistral says 209M Timestral leads on three benchmarks
- Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3
- Is ExlService Holdings (EXLS) Undervalued As Agentic AI Expands Across Insurance And Healthcare?
- AI trade is at a 'critical' moment as rising rates threaten a capex slowdown
- Dito Joins the Google Cloud Managed Security Services Partner Program, Bringing AI-Powered Security Operations to Enterprise and Public Sector
- Tenstorrent raised $1.5 billion, more than doubling its valuation to $6.2 billion
- Tufin Advances AI-Ready Network Security with Expanded Platform Coverage and Compliance in TOS 5.4
- Jim Cramer says don't chase the AI rebound just yet. Here's what needs to happen first
- Chip Stocks Rally, SanDisk Surges Over 7%: Has the AI Hardware Stock Correction Ended?
- Six AI Real Estate Plays That Go Beyond Data Centers
Comments
Please log in to post a comment.