Ben Thompson, founder of Stratechery, argued in a recent interview that Nvidia’s exceptional profitability may be less durable than it appears as the AI spending cycle enters a more contested phase. His view centers on two pressure points. First, he said Nvidia has supported newer cloud providers, or “Neoclouds,” through equity stakes and roughly 25% backstops tied to commitments to keep buying Nvidia compute through 2030. That may help sustain GPU shipments, but Thompson said the risk does not disappear; it shifts back onto Nvidia if compute demand weakens or those buyers cannot keep purchasing. In his framing, that amounts to a hidden reduction in profit and functions like an indirect price cut. Second, Thompson said Nvidia’s energy-efficiency edge matters most when power is scarce. He argued that unexpectedly resilient U.S. electricity supply over the past two years — including natural gas generation in West Texas, restarted nuclear plants, and grid-related deployments by Elon Musk — gives hyperscalers such as Amazon and Google more time to improve in-house chips like Trainium and TPU. That, in turn, could erode Nvidia’s technical moat. Even if the current AI boom ends in oversupply and a market correction, Thompson said the resulting buildout of power infrastructure may still become the most durable legacy of the cycle.

U.S. stocks fell for a third straight session Tuesday as higher long-term bond yields put fresh pressure on richly valued technology names. The Nasdaq Composite dropped 1.33%, underperforming the Dow Jones Industrial Average and the S&P 500, while the 30-year U.S. Treasury yield briefly touched 5.338%, its highest level since 2007. The move was part of a wider global bond sell-off that also pushed long-dated yields higher in France, Germany, Japan and the U.K. Markets are increasingly focused on the growing debt burden tied to artificial intelligence expansion. According to figures cited in the report, AI-related bond issuance has reached $489 billion so far this year, well above an earlier full-year 2025 estimate of roughly $322 billion, while The Wall Street Journal reported that nine major technology companies have about $3 trillion in off-balance-sheet AI commitments. That backdrop hit semiconductors, memory, optical communications and AI cloud-service providers especially hard. Investors are also weighing fiscal deficits, oil-driven inflation risks tied to the Iran situation, and a heavy event calendar that includes U.S. tariffs on some Canadian products, a 20-year Treasury auction, Federal Reserve minutes and China’s one-year LPR decision.

A summer heatwave could erase about 1% of the European Union’s 2026 GDP, or roughly €180 billion, according to estimates cited from Dutch bank Triodos Bank. The report says the biggest hit comes from weaker labor productivity as extreme heat cuts working hours and halts activity in outdoor construction and logistics. France is seen as the hardest-hit economy, with GDP potentially reduced by 1.4% and full-year growth turning into a 0.6% contraction, while the Netherlands could see growth nearly stall. Triodos breaks the damage into four channels: labor losses, tighter food supply and lower dairy output, power disruptions and higher electricity prices, plus transport bottlenecks across land and inland waterways. The report also points to low water levels in rivers including the Seine, Rhine and Danube, pressure on nuclear generation, and weaker crop output for corn and sunflower. Analysts cited in the piece say these are supply-side shocks that conventional rate cuts or fiscal stimulus cannot easily offset. The article also highlights inflation risks for the European Central Bank, wildfire losses, heat-related excess deaths, and a possible shift in summer tourism from southern Europe to cooler northern destinations. Triodos chief economist Hans Stegeman warned that climate damage is no longer a distant economic risk and argued that cutting the cost requires slowing climate change itself.

Bank of America Securities chief strategist Michael Hartnett said the preferred approach in the current AI bubble is to stay long dominant AI-related technology stocks while also buying assets that have been neglected for a long time, and to short AI-linked bonds. In his latest note dated Aug. 15, Hartnett argued that this mix could capture two-way returns during what he described as the final melt-up phase of a nominal GDP bubble. BofA’s bull-and-bear indicator slipped from 9.7 to 9.3, but it remains in an extremely bullish zone and still carries a sell signal. Even so, global equities have continued to rise since that signal was issued in May. The report also said flows are moving structurally into gold and commodities, while tech stocks just saw their biggest weekly outflow in seven weeks. Private client equity allocation climbed to a record 66.4%, while cash fell to the lowest level on record and bond exposure dropped to its lowest since 2022. Hartnett also pointed to more than $1 trillion in AI capital spending combined with negative cash flow as a source of heavy issuance pressure for related bonds, and said U.S. midterm elections in November are a key political variable for what comes next.

Matthew Sigel, head of digital assets research at VanEck and manager of the VanEck Onchain Economy ETF (NODE), said the current AI infrastructure boom should not be viewed as a replay of the 19th-century U.S. railroad bubble. Speaking on The Rollup podcast episode “AI Super Cycle,” aired on Aug. 10, 2026, Sigel argued that the key difference lies in financing: railroad expansion relied on government-led land grants and speculative bond issuance, while today’s AI buildout is backed by private-sector contracts, multiyear leasing commitments, customer prepayments, and more than $2 trillion in cloud backlog held by the four largest cloud providers. He added that AI factories can begin producing value once connected to power, fiber, and chips, unlike railroads, which required a completed coast-to-coast network before their utility fully emerged. Sigel also said crypto’s weak price action has less to do with macro conditions and more to do with institutions losing conviction in major layer-1 networks such as Solana and Ethereum. VanEck has cut exposure to mainstream L1s since the U.S. election, he said, after many tokens doubled without a comparable acceleration in real adoption or breakout applications. In their place, the firm has turned more attention to enterprise chains linked to companies including Circle, Stripe, Robinhood, and, as Sigel noted, even research efforts at Wells Fargo. He said regulated institutions want predictable fee structures and are reluctant to place meaningful capital directly on open public chains. Sigel said NODE has outperformed Bitcoin by nearly 100 percentage points over the past 15 months, driven largely by an early bet on Bitcoin miners pivoting toward AI data center infrastructure.

U.S. retail sales fell more than expected in July, adding to concerns that consumer demand may be losing momentum in the second half of the year. Data released Friday by the U.S. Census Bureau showed retail sales dropped 0.6% month over month, the steepest monthly decline since May 2025, versus market expectations for a 0.1% increase. The pullback was led by weaker vehicle purchases and softer online retail activity. Excluding autos and gasoline, retail sales slipped 0.2% in July, missing forecasts for a 0.3% rise. The closely watched control-group measure, which feeds directly into GDP calculations through personal consumption expenditures, fell 0.4% from the prior month. Economists and market participants are paying particular attention to that reading because it is widely used as a gauge of underlying consumer strength. The report also pointed to several short-term drags on July spending, including the timing shift of Prime Day-related promotions into June, weaker apparel, furniture and gasoline sales, hot weather around the July 4 holiday, and fading post-World Cup demand. On a yearly basis, retail sales still rose 5.0%, but the pace slowed. Bank of America card data cited in the report showed lower-income households have recently outpaced higher-income households in consumption growth, suggesting the long-running K-shaped spending split may be narrowing.

Google has released Gemini 3.7 Flash on Aug. 13, shortening its model update cycle to roughly three weeks after Gemini 3.6 Flash. According to Google’s announcement, the new model is positioned as a high-value offering for coding and agent tasks, while also targeting document-heavy knowledge work and web development. It supports text, image, audio, and video input, comes with a 1 million-token context window, and can generate up to 64,000 output tokens. Google said Gemini 3.7 Flash improved on several benchmarks versus Gemini 3.6 Flash, including FrontierCode 1.1, which rose from 34.4% to 43.6%, AutomationBench from 17.0% to 30.4%, and the document-understanding benchmark GDP.pdf from 22.0% to 34.0%. The product is being offered through API and enterprise channels, including Gemini API, Google AI Studio, Antigravity, Android Studio, and Gemini Enterprise. Consumer access is available through Gemini Spark under AI Pro and Ultra plans. Google is not releasing open-weight access for the model. Pricing is a central part of the launch. Through Dec. 31, 2026, input costs are set at $0.75 per 1 million tokens and output at $3.75, before rising to $1.5 and $7.5 in 2027. Using an 80/20 input-output mix, ABMedia estimated blended cost at about $1.35 per 1 million tokens, below Sonnet 5 at $3.60 and GPT-5.6 Terra at $4.00.

The White House Office of Trade and Manufacturing Policy on Aug. 13 released a 25-page report titled The Great Transshipment Scam, alleging that China uses more than 40 countries to reroute goods and avoid U.S. tariffs. The report estimated the annual scale of the practice at between $40 billion and $303 billion, but it also said those figures come from five separate methodologies and should neither be added together nor directly compared. Taiwan was placed in Tier 1, alongside Japan, South Korea, Canada, the European Union, India, Israel, and Mexico. The report described that group as large, diversified export platforms where transshipment risk is embedded within substantial volumes of legitimate trade. It also outlined economic impact estimates based on a $75 billion midpoint scenario, including a $113 billion to $150 billion annual GDP hit, displacement of 450,000 jobs, and $19 billion to $26 billion in lost federal tax revenue. The document also referenced a proposed AI-based enforcement system called Detective Border and said it is still too early to judge the net effect of the administration’s tariff and anti-transshipment measures.