Saturday, May 16, 2026
☀️ A golden retriever somewhere just discovered a puddle and is about to make it its whole personality. Channel that energy today.
May 15, 2026 — 4:00 PM ET close (Markets closed Saturday)
Broadcom fell sharply Friday as investors rotated out of semiconductor and AI infrastructure plays following the Trump-Xi meeting's failure to produce major trade deals. The chip designer, which has been a key beneficiary of AI capex cycles, faced selling pressure as geopolitical tensions and stalled negotiations raised questions about near-term demand visibility. The broader semiconductor sector weakness reflected profit-taking after a strong April rally that had pushed valuations to elevated levels.
The Strait of Hormuz remains effectively closed as US-Iran peace negotiations stalled, with Trump warning Iran to reach a deal or face 'annihilation.' The blockade has disrupted roughly 4 million barrels per day of crude and fuel flows, pushing Brent crude to $109 (up 52% YoY) and triggering a 17.9% YoY surge in energy prices that drove April CPI to 3.8%. The International Energy Agency warned Friday that even if fighting ends next month, the global oil market could remain materially undersupplied through October due to inventory depletion and tanker traffic constraints. This is the structural driver of Friday's selloff: without a near-term resolution, energy inflation will persist, forcing the Fed to keep rates elevated, which crushes growth expectations. The failed Trump-Xi meeting was supposed to be a circuit-breaker on this risk, but instead it confirmed that the Middle East conflict will remain a persistent drag on growth and inflation for quarters to come.
The April Consumer Price Index released Tuesday showed headline inflation at 3.8% YoY—the highest since May 2023—with energy costs surging 17.9% YoY as the Strait of Hormuz blockade persists. Core CPI (excluding food and energy) rose 2.8% YoY, still well above the Fed's 2% target. The hot print has forced a dramatic repricing of Fed expectations: markets now price in virtually zero probability of rate cuts through end of 2026, with only 8% odds of a cut by June 30, 2027. The 10-year Treasury yield surged to 4.46%, just 2 bps shy of its 2026 high. This is the critical inflection point for markets: the soft-landing narrative (slowing inflation, resilient growth, Fed cuts) has collided with stagflation reality (sticky inflation, slowing growth, Fed holds). The April CPI print is the smoking gun—it proves that inflation is not transitory and that the Fed's 75 bps of cuts last year may have been premature.
The Russell 2000 fell 2.44% Friday, nearly double the S&P 500's 1.24% decline, as investors fled small-cap growth stocks that are most vulnerable to higher discount rates and slower growth. Small-cap companies—which tend to be less profitable and more leveraged than large-caps—are priced on the assumption of strong future growth and lower interest rates. When the Fed signals rates will stay elevated through 2027, small-cap valuations compress sharply. The Russell 2000 is now down 1.85% YTD, significantly underperforming the S&P 500's +5.49% gain, reflecting a structural rotation from growth to value and from small-cap to large-cap. This is a classic risk-off signal: when investors fear recession, they flee unprofitable small-caps and rotate into large-cap defensive stocks with stable cash flows. The Russell's weakness Friday suggests that institutional money is repricing growth expectations downward and positioning for a slower-growth, higher-rate environment.
Broadcom fell 3.87% Friday as the semiconductor sector faced broad-based selling after a strong April rally. The chip designer, which has been a key beneficiary of AI capex cycles, faced profit-taking as investors rotated out of growth stocks on stagflation fears. Broadcom's weakness is particularly significant because it's a bellwether for AI infrastructure spending: if Broadcom is selling off, it suggests that institutional investors are questioning the sustainability of AI capex cycles. The failed Trump-Xi meeting also matters—investors had priced in potential Chinese demand for US chips if trade tensions eased, but the lack of a deal removes that upside catalyst. Broadcom's 3.87% decline is modest compared to the Russell 2000's 2.44% plunge, but it's directional: it signals that even the most profitable AI-beneficiary companies are not immune to broader risk-off sentiment.
President Donald Trump and Chinese President Xi Jinping met in Beijing on May 15 but departed without major trade agreements or progress on resolving the Middle East conflict. Markets had priced in optimism that Trump would negotiate a tariff rollback or at least secure Chinese cooperation on Iran, but the summit produced only vague statements about keeping the Strait of Hormuz open. The failure to achieve concrete results triggered a sharp selloff: the S&P 500 fell 1.24%, the Russell 2000 plunged 2.44%, and the Nasdaq dropped 1.54%. The immediate cause is straightforward—investors had front-run a deal, and the absence of one forced a repricing. But the deeper issue is structural: without a breakthrough on Iran, the blockade of the Strait of Hormuz persists indefinitely, keeping crude oil elevated (Brent at $109, up 52% YoY) and inflation sticky at 3.8% YoY. This forces the Fed to maintain restrictive policy through 2027, crushing growth expectations and leaving equities trapped between recession risk and persistent price pressures. The 2s/10s curve compressed to 48 bps Friday—a classic stagflation signal—as investors repriced growth downward while inflation expectations remained anchored above 3.5%. For markets, the Trump-Xi meeting was supposed to be a circuit-breaker on geopolitical risk; instead, it confirmed that the Middle East conflict will remain a persistent drag on growth and inflation for quarters to come.
💡 The Strait of Hormuz is a critical chokepoint through which roughly 20% of global oil passes daily. When blockaded, it creates supply constraints that push oil prices higher, which flows into inflation, which forces central banks to keep rates elevated. This creates a vicious cycle: higher rates slow growth, but inflation stays hot, leaving policymakers with no good options.
Bill Ackman's Pershing Square Capital disclosed a significant position in Microsoft on Friday, purchasing shares at a valuation of 21x forward earnings—broadly in line with the S&P 500 multiple and well below Microsoft's historical average. Ackman's move is a contrarian bet: Microsoft fell 4% on its Q1 earnings despite beating estimates, as investors worried that AI could disrupt its core software business. The purchase signals that Ackman sees the selloff as overdone and believes Microsoft's Azure cloud division and partnership with OpenAI position it to capture AI infrastructure spending. This is a 2nd-order bet on AI adoption—not on chip makers or model developers, but on the cloud platforms that will host and monetize AI workloads. The timing matters: Ackman is buying into weakness after the Trump-Xi meeting spooked growth stocks, suggesting he views the current pullback as a capitulation moment rather than a structural repricing.
💡 Forward earnings multiple (P/E) — the stock price divided by expected earnings over the next 12 months. A 21x multiple means investors are paying $21 for every $1 of expected future earnings. Microsoft's historical average is higher, so Ackman is arguing the stock is now cheaper relative to its own history.
Alphabet delivered a blockbuster April, posting a 34% monthly gain—its strongest since 2004—on the back of Q1 earnings that beat across cloud, advertising, and Waymo autonomous driving. Google Cloud is experiencing explosive demand as enterprises adopt Gemini AI models, while Waymo's autonomous taxi service is expanding in multiple cities. Most significantly, the market is now giving Alphabet credit for its custom TPU (Tensor Processing Unit) chips, which are emerging as a legitimate alternative to Nvidia's GPUs for AI workloads. This is a structural shift: for years, Nvidia had a near-monopoly on AI chip supply, but Alphabet's TPUs are proving competitive on price and performance for certain workloads. Alphabet's market cap now exceeds Apple's, making it the second-most valuable company globally. The April rally reflects a broadening of the AI narrative beyond pure chip makers to the companies that will own the AI infrastructure stack—cloud platforms, custom silicon, and AI-native applications.
💡 TPU (Tensor Processing Unit) — a custom chip designed specifically for AI workloads, optimized for matrix multiplication and neural network training. Unlike GPUs (which are general-purpose graphics processors), TPUs are purpose-built, which can make them faster and cheaper for certain AI tasks.
Meta Platforms beat Q1 earnings estimates but immediately spooked investors by raising 2026 capex guidance to $125-145B—a significant increase from prior expectations. The company is doubling down on AI infrastructure, building massive data centers to train large language models and power its AI-driven recommendation systems. While the earnings beat was solid, the capex raise signals that Meta's profitability gains may be temporary if spending accelerates faster than revenue growth. The stock fell 9% on the news, reflecting investor concern that the AI capex cycle could persist longer than expected, pressuring margins. This is a critical inflection point for the Magnificent 7: companies are now in a race to build AI infrastructure, and the winner will be whoever can monetize that infrastructure fastest. Meta's bet is that AI-driven recommendations will drive higher ad engagement and pricing power, but that's a 2-3 year thesis, not a near-term earnings driver.
Bitcoin fell 2.79% to $79,105 on Friday, extending losses from the broader market selloff triggered by the failed Trump-Xi meeting and April CPI data showing inflation at 3.8% YoY. Ethereum dropped 3.42% to $2,847, while Solana fell 4.31% to $86.14, underperforming Bitcoin as investors rotated out of higher-beta altcoins. The crypto selloff reflects the asset class's tight correlation with growth stocks and risk appetite: when equities fall on stagflation fears, crypto follows because both are priced on the assumption of lower discount rates and higher growth. The failed geopolitical breakthrough also matters—investors had priced in a potential resolution to the Middle East conflict, which would lower oil prices and inflation, which would allow the Fed to cut rates. Without that breakthrough, the stagflation scenario becomes more likely, which is bearish for risk assets including crypto. Bitcoin's failure to hold above $80K is technically significant, as the level had been a key support.
💡 Correlation — the degree to which two assets move together. Bitcoin and the S&P 500 have a correlation of ~0.6-0.7 in 2026, meaning they tend to move in the same direction. When growth stocks fall on recession fears, Bitcoin typically falls too, because both are 'risk-on' assets.
Despite Solana's 4.31% decline Friday, spot Solana ETFs recorded $39.2M in net inflows last week—the largest weekly inflow since February—signaling that institutional investors are buying the dip. The inflows are notable because they contradict the price action: typically, when an asset falls sharply, ETF flows turn negative as retail investors panic-sell. Instead, the positive flows suggest that institutions view the current weakness as a buying opportunity. Solana's narrative has shifted from pure speculation to technical fundamentals: the Firedancer validator client upgrade has dramatically improved network performance, developer migration to Solana is accelerating (the chain ranked second only to Ethereum for new developer inflows in 2025), and the launch of staking-enabled Solana ETFs in October 2025 has created a structural bid from traditional finance. The $39.2M inflow is modest in absolute terms, but it's directional—it shows that the institutional bid for Solana remains intact despite near-term price weakness.
A groundbreaking study published this week in *Nature Neuroscience* reveals that octopuses possess taste receptors distributed throughout their eight arms, allowing each limb to independently detect and evaluate food without input from the central brain. Researchers at Woods Hole Oceanographic Institution observed that octopus arms would reject or accept food based on chemical signals detected directly in the arm tissue, even when the central nervous system was not involved in the decision. This is a radical departure from how humans and most animals process sensory information—we rely on centralized brains to integrate signals from distributed sensory organs. Octopuses, by contrast, have evolved a decentralized sensory architecture where each arm is semi-autonomous, capable of tasting, touching, and even learning independently. This explains why octopuses are so remarkably intelligent and adaptable: they're essentially nine brains (one central, eight distributed in the arms) working in parallel, each capable of solving problems and making decisions. The discovery has profound implications for understanding how nervous systems can be organized and how intelligence can emerge from distributed rather than centralized processing.
💡 Chemoreceptors — sensory cells that detect chemical signals (like taste and smell). Most animals have chemoreceptors concentrated in the mouth, nose, and tongue. Octopuses have them distributed throughout their arms, making each arm a semi-independent sensory organ.