Crypto bettors see little near-term risk of an AI crash in the U.S., even as warning signs emerge from China. On Polymarket, the Polygon (POL)-based prediction market, the chance of an AI-industry bubble burst is priced at only 9%.
Summary written by Upside AI from the original article. Read the source for full details.
Read the original on Yahoo Finance ↗
Keywords
artificial intelligenceAI bubbleNVDAQQQChina
Open a keyword to see other news on the same topic.
Related tickers and likely impact
NVDA
▼ Likely negativeNvidia
Relevance: medium
| Market | US stocks |
|---|---|
| Sector | Information Technology |
| Industry | Semiconductors |
| Market cap | $5.1T |
| Close on 10/26/2026 | 215.98 |
| Change that session | +2.38% |
QQQ
▼ Likely negativeInvesco QQQ Trust
Relevance: low
| Market | US stocks |
|---|---|
| Sector | ETF |
| Close on 10/26/2026 | 712.11 |
| Change that session | +1.09% |
SPY
▼ Likely negativeSPDR S&P 500 ETF
Relevance: low
| Market | US stocks |
|---|---|
| Sector | ETF |
| Close on 10/26/2026 | 765.22 |
| Change that session | +0.69% |
Impact is scored by AI from the headline and may be wrong. It is not investment advice.
Related news
- Google's Ex-CEO Says China Trails US by 6 Months in AI Race, but Could Still Win Due to This Strategy: 'We're Already Losing'
Former Google CEO Eric Schmidt said on a podcast that China trails the U.S. in the AI race by about six months, but could still win because its strategy offers products for free. He argued that Chinese AI models are often free or cheaper, which could attract users in developing economies, and credited China's progress partly to chips he said were pilfered or smuggled into the country, as well as to distillation of Western AI systems. Schmidt rejected comparisons between AI and a speculative bubble, pointing to Anthropic's revenue growth and paying customers as evidence of real demand. The article also cites IMD professor Howard Yu, who said the U.S. lead could prove perishable since many users do not need the most advanced models, and Polsia AI CEO Ben Cera, who said the bubble lies in the price of AI rather than AI itself.
- Michael Burry Is Bearish on Palantir, Which Is Up 19% Over the Past Year, but His Oracle Short Just Got a Boost From a 53% Stock Drop. Is His AI Bubble Bet Still Worth Following?
Michael Burry, who made his fortune betting against the U.S. housing market before the 2008 crisis, remains bearish on AI overinvestment even after disbanding Scion Capital. He holds put options on Nvidia, Palantir and Oracle, but so far only the Oracle bet has paid off, while Nvidia and Palantir have not fallen as he expected. Burry argues the market is in denial and propped up by borrowed money, and says this denial phase lasted six to nine months before previous crashes. He has also criticized Nvidia’s chip depreciation math, while buying MercadoLibre and Deckers Outdoor. The author concludes that AI investors should still follow this bear case to maintain an objective view.
- S&P 500 Just Hit Record High in October. History Stands Ready to Repeat Itself.
The S&P 500 is up 14.0% year to date and hit a record on October 6. Bianco Research says its 41 AI-related companies make up 48.68% of the index's total market cap, led by Nvidia at $5.5 trillion. These companies are spending heavily on infrastructure: Microsoft put $115.95 billion into fiscal 2026 capex, Meta guided to $130–145 billion, and Alphabet's long-term debt rose from $46.5 billion to $98.2 billion while it suspended buybacks. The article compares this concentration to the 19th-century railroad boom, when overbuilding ended in panics, and argues AI safety risks could become index-wide risks. BTIG notes more stocks hit 52-week lows than highs, a pattern it compares to 2000.
- AI FOMO driving some family offices into tech investments, survey suggests
A survey of 155 North American family offices by RBC Wealth Management and Campden Wealth found that 85% rank AI as their top investment theme, even though 75% expect an AI bubble to burst within five years. RBC's Bill Ringham said fear of missing out drives a short-term focus: 59% named AI the top theme of 2026, but over the two-to-five-year horizon, 69% favored U.S. equities versus 64% for AI. The most-cited market risks were excessive government borrowing (76%), an AI bubble bursting (75%) and AI advancements affecting employment (66%). Among those planning changes, 31% intend to increase direct investments in private equity, while just over a fifth plan the same for real estate and venture capital.
- China's AI bubble bursts, $480 billion in market value evaporates in under 4 months
The STAR50 index, focused on AI and semiconductor companies, has fallen 40% in under 4 months, wiping out about $480 billion in market value, while the CSI 300 is down about 20% from its June peak. Many high-flying stocks have lost about half their value: Unitree has halved since its August listing, and MetaX and Moore Threads are also down about half since July. Optical component makers for data centers, Zhongji Innolight and Eoptolink, have dropped about 60% from their peaks, partly due to moves in Washington to ban their equipment from U.S. data centers. Even after the decline, STAR50 stocks trade at about 100 times trailing 12-month earnings, well above roughly 30 times for the Nasdaq 100, leaving many investors worried the sell-off is not over. Some funds still view the market positively, seeing lower prices as an opportunity for cautious investors.
- Nvidia CEO Jensen Huang Says US Must Take China ‘Seriously,’ Warns Against Underestimating Its AI and Chipmaking Prowess: ‘If There's One Country That Could Manufacture Things at Scale’
At Nvidia's GTC Berlin conference, CEO Jensen Huang said the US must take China seriously as a rival in chips, AI and research, calling it the one country that could manufacture things at scale. He argued that American frontier labs, computing systems and infrastructure remain the strongest in the world. According to an IDC analysis, Nvidia held 34% of China's AI accelerator market in the first half of 2026 versus 28% for Huawei, but its shipments there fell 44% year-on-year to 678,000 accelerators. Earlier, Huang urged the US to embrace rather than ban China's open-source AI models, arguing they could expand AI adoption without threatening closed-model providers such as OpenAI and Anthropic.
Related markets
Latest news
- Lion Rock Receives State and County Permits for Pegmatite Extraction & Bulk Sampling
Lion Rock Resources received a state mining license from South Dakota's Department of Agriculture and Natural Resources (DANR), which together with a previously approved Lawrence County Conditional Use Permit completes permitting to mine pegmatite and conduct bulk sampling at its Volney project in the Black Hills. The company is advancing metallurgical studies, targets a bulk sampling program in 2027, and is evaluating small-scale mining that could generate revenue. Recent drilling returned tantalum intercepts of up to 3,715 ppm Ta₂O₅ over 1.0 m, alongside tin and lithium mineralization. No production decision has been made, and historical concentrate grades of up to 57% Ta₂O₅, 62.5% Sn and 6.3% Li₂O have not been fully verified by a Qualified Person.
- Several Important Factors Protect Amazon’s Margins and Secure Its AI Moat
The author says they added to an Amazon position at $249.09, up 7.92% year to date, as the business accelerates. The article argues that in-house chips, the Bedrock platform supporting multiple AI models and automated logistics help protect margins. Amazon's chips business is reported to run at an annual revenue rate above $25 billion, while AWS posted a 39% operating margin on $42.2 billion in revenue, up 36.7%. The AWS backlog stands at $496 billion, and operating income rose 43.24% to $27.46 billion while revenue grew 19.62%. The article also flags risks: 2026 capex is expected at about $220 billion, trailing 12-month free cash flow was -$7.6 billion, and Q2 GAAP EPS included a $53.4 billion pre-tax gain on the Anthropic stake. Q3 operating income guidance is $22.5 billion to $26.5 billion, compared with $17.4 billion a year earlier.
- CoreWeave Vs. Nvidia: Demand For Silicon is Not The Same Thing as Solvent, Credit-Backed Demand for Physical Infrastructure
CoreWeave and NVIDIA both posted record quarters on AI demand, but the article contrasts NVIDIA's cash from chip sales with CoreWeave's borrowing to build rented data centers. NVIDIA reported fiscal Q2 revenue of $96.221B, up 105.8%, with a 75.0% non-GAAP gross margin, while CoreWeave's revenue rose 112.3% to $2.575B alongside a $104.2 billion backlog. CoreWeave's interest expense rose to $640M from $267M, and its free cash flow was -$5.743B. NVIDIA now offers minimum revenue guarantees, with obligations limited to $108.5B, to help NeoClouds secure financing. The article notes CoreWeave shares fell 36.99% over the past month and are down 22.93% year to date, while NVIDIA is up 16.09%, and argues NVIDIA is better positioned because it is paid at the hardware sale.
- Amkor Technology Halved Over The Last 6 Months: Two Chip Analysts Say 145% Returns Await Those Who Buy Now
Amkor shares, at $36.81, have fallen 51.18% over six months after a Q3 miss: EPS of $0.76 came in below the $0.7766 consensus, and revenue of $1.99 billion missed roughly $2.04 billion expected. Revenue growth slowed to 0.18% year over year, the Communications share of revenue fell to 37% from 51%, and Q4 EPS guidance of $0.65 to $0.75 sits below the $0.7774 consensus. Two analysts, Needham's Charles Shi and UBS's Randy Abrams, carry $90 targets set before the miss, implying about 144% upside from current levels, while the $74.50 consensus target implies about 102%. The article cites AI packaging demand and Q3 gross margin expansion to 18.3% from 14.3%, but notes roughly $2.5 billion in full-year capex. The stock is down 6.3% year to date, while the S&P 500 is up 12.24%.
- Prediction: Broadcom Stock Could Be Worth Twice as Much by 2030
Broadcom's AI semiconductor revenue reached $16.70B, up 221% year over year, and total revenue rose 85.5% to $29.59B, yet the stock is down 6.45% this year at $323.24. The article says shares sit near the 52-week low of $288.95 rather than the high of $493.28, and points to competition from Google's cheaper second sources and a 210 basis point sequential drop in gross margin. The article's model sets a one-year target of $402.33, while the consensus target is $527.59. Reaching $650 by 2030 would require a 101.1% gain, or about 35x the forward EPS of $18.376. The article ties this to AI revenue of roughly $115 billion in fiscal 2027 and $230 billion in fiscal 2028, with management saying fiscal 2028 EPS should exceed $30. The main risk noted is that a few AI labs drive most of the demand.
- Microsoft Is Paying Workers to Leave So It Can Pay for AI
Microsoft offered voluntary buyouts on October 26 to about 60,000 employees, roughly 27% of the 223,000 full-time staff as of June 30. The article links the offer to rising AI spending, though the announcement does not mention AI or cost cutting. The stock rose 1.54% to $532.60, and the company is valued at about $3.95 trillion. Fiscal 2026 capex reached $115.95 billion, up 79.62%, while free cash flow fell 6.46%, and calendar-2026 capex is put at about $175 billion. The offer covers level 70 and below on selected teams such as Xbox, Windows, Copilot and Finance, while data center, sales and marketing staff are largely excluded. Wells Fargo, which rates the stock Overweight with a $725 target, estimates that if one-third of eligible workers accept, operating margin could rise about 2 percentage points a year, though uptake and total cost were not disclosed.
Get the briefing on Telegram or Zalo
Join the Upside Telegram channel or Zalo group for the Vietnam stocks and gold briefing at 8:00 and 19:00 (Vietnam time) on trading days, plus a weekly wrap on Saturday morning.