Last week’s modest decline, beneath the recent discord, wasn’t material to change our current fundamental indicators – Sowell’s technical gauges remain composed and fully invested.
A Moment of Market Discord – Hawkish Fed and AI Hangover
Well, the second week of December was truly a class in market contradiction, leaving both stock and bond investors scratching their heads. The headliner was a perfectly executed maneuver by the Federal Reserve, which, with its characteristic professional wit, delivered what can only be described as a "Hawkish Cut."
The Dual Mandate & The Bond Market Enigma
The week’s primary event was the Federal Open Market Committee (FOMC) meeting, where the Fed delivered the anticipated 25-basis-point cut to the policy rate. This move was justified by Chairman Powell's assessment: "Conditions in the labor market appear to be gradually cooling, and inflation remains somewhat elevated." The statement perfectly sums up the core struggle—a dual mandate where the employment side is showing weakness while inflation stubbornly persists.
The bond market's reaction, however, was counterintuitive to a rate cut. Treasury yields rose overall, pushing the prices of existing bonds down. This surprising move was due to nuances in the Fed's outlook and perceptions of underlying economic strength. The yield curve steepened as short-term yields were anchored by the cut, but longer-term yields—driven by inflation fears and the limited future cuts signaled in the "dot plot"—pushed higher. This steepening signaled that, despite the immediate policy easing, the market remained wary of the long-term inflation trajectory.
The Stock Market's AI Cooling
The equity market, represented by the S&P 500 Index, ultimately closed the week down 0.61%. While the immediate post-Fed rally was a flash of bullish exuberance, the weight of a few select sectors proved too heavy.
The culprit? The ever-powerful, but currently temperamental, technology sector. Communication Services and Technology stocks were the primary drag, led by heavyweights like NVIDIA, Google, Broadcom, and Meta. This pressure was the market’s latest demonstration of the familiar boom-and-bust cycle. Any new "shiny object"—from the internet boom and cannabis to crypto, EVs, and now AI—comes with its own journey through exuberance, high expectations, disappointment, and rationalization.
Investors are realizing that even with a transformational technology like AI, not every darling can maintain a trillion-dollar valuation based solely on hype. In a twist of irony, the year-to-date performance champions among the top 10 stocks are actually Google and Broadcom, not NVIDIA, demonstrating the complexity of sector leadership.
An Economic Mixed Bag
The week's economic data only reinforced the ambiguity facing the Fed:
NFIB Small Business Optimism Index: Rose to 99.0 in November, slightly above expectations. However, the most telling detail was the jump in the net percent of business owners raising average selling prices (up 13 points to 34%—the highest since March 2023). Inflationary pressures are clearly being passed through. Meanwhile, the net percent expecting better business conditions declined by 5 points, suggesting optimism is fragile.
Labor Sentiment: The data was a true mixed bag. Employment Cost Index (ECI) cooled slightly (+0.8% vs. +1.0% estimated), a positive signal for inflation. However, Initial Jobless Claims jumped to 236K (above the 220K estimate), hinting at some weakness, while Continuing Claims dropped significantly to 1.838M, suggesting those who lose a job are quickly finding a new one. This supports the Fed’s observation of a "gradually cooling" labor market.
Leading Indicators: Continued to suggest caution, coming in at -0.3%, matching the prior month.
A Nod to the Old Guard and Trade Policy
Amid the volatility of new-age tech stocks, the ultimate safe-haven asset, old-fashioned gold, posted a weekly gain of 2% and substantial YTD appreciation, reportedly up 60% in 2025. This performance underscores investor anxiety, showing a clear appetite for hard assets amidst central bank uncertainty and geopolitical risk.
Finally, the US December Trade deficit of -$52.8 billion continues to narrow, hitting its lowest level since June 2020. This trend, a key policy goal of the current administration, is often attributed to President Trump's tariff policy, which has reshaped global trade flows and domestic production incentives.
Week Ahead
With the drama of the Fed decision and the AI sector's reality check now behind us, the financial spotlight immediately shifts to a barrage of high-impact economic releases next week. If last week was the monetary show, the week of December 15th will be the data deluge.
The market's current narrative—that the Fed is cutting rates but inflation remains a problem—will be severely tested. Key releases include the all-important Consumer Price Index (CPI) on Thursday, which will tell us exactly how stubborn inflation really is, and Tuesday's Unemployment Rate and Nonfarm Payrolls (consolidated for October and November after reporting delays), which will provide the clearest picture yet of the "gradually cooling" labor market Powell described.
Furthermore, the market will digest the ADP Employment Change as a warm-up act and continue to monitor weekly Jobless Claims for confirmation of the cooling trend. Though the PCE (the Fed’s preferred inflation gauge) is scheduled later in the month, the upcoming data provides more than enough fodder to drive bond yields and test the resolve of the S&P 500, ensuring the holiday cheer remains tempered by economic uncertainty.
"One perhaps self-serving observation. I’m happy to say I feel better about the second half of my life than the first. My advice: Don’t beat yourself up over past mistakes – learn at least a little from them and move on. It is never too late to improve. Get the right heroes and copy them. You can start with Tom Murphy; he was the best.”
— Warren Buffett, Thanksgiving Message, Nov 10, 2025.
Fork in the Road in the AI Hardware Race
By Fiona Zhang
Google’s home-grown AI chip, the Tensor Processing Unit (TPU), has quietly graduated from cool internal toy to possible market plot twist. What started as a bespoke accelerator built to run Google’s own AI brainchildren is now popping up in headlines and, more intriguingly, in the tape. And investors are definitely watching.
Lately, whispers that several big AI players are testing or expanding workloads on TPUs have been enough to send traders scrambling. Alphabet catches a bid on those headlines, while Nvidia and AMD see the occasional wobble as the market remembers that the GPU throne might not be welded in place forever. These aren’t seismic rotations yet, more like tremors, but they reveal a market hypersensitive to any hint that AI computing could someday be a multiplayer game.
Part of it is pure capex physics. If even a slice of training or inference workloads shifts toward Google’s cloud-hosted TPUs, that means fewer GPUs purchased elsewhere, and more long-term revenue chugging through Google Cloud. Analysts have already pointed out that faster TPU adoption could brighten Google Cloud’s growth trajectory while casting a bit of shade on pure-play chipmakers.
That’s why even a rumor that a major social platform is “testing TPUs at scale” can vaporize or create billions in market cap in minutes. And you could see that tension around Nvidia’s latest earnings, too: they delivered another beat, yet the stock still finished the day in the red. In this environment, it’s not enough for Nvidia to surprise; they have to surprise the surprise, because competition fear is now part of the trade.
But this isn’t some tidy duel where one chip simply replaces another. GPUs and TPUs aren’t substitutes so much as specialists: GPUs are the general-purpose heavy lifters with massive parallelism; TPUs are laser-focused sprinters designed for the kinds of matrix math that fuels deep learning.
All of this sits atop Nvidia’s deeply entrenched software ecosystem, which remains sticky, beloved, and very hard to replace. No one expects customers to abandon GPUs overnight. Most will run hybrid fleets with GPUs doing some jobs and TPUs doing others, because switching costs are painful and CUDA (Compute Unified Device Architecture) still rules the AI software universe.
Meanwhile, Google still has to prove it can supply, support, and scale TPUs globally across multiple hardware generations — a very non-trivial hurdle.
Still, markets are telling their own story. For the first time in years, investors can imagine a world where AI compute demand isn’t captured by a single architecture. Even a little diversification in the supply chain could reshape cloud capex, ripple across semiconductor names, and give Alphabet a far bigger footprint in the economics of AI.
For now, TPUs sit in an interesting middle zone: not yet hitting the trend’s peak, but no longer a residual squeak. What they are is a new variable for investors to watch — a signpost indicating whether the AI hardware race remains a one-horse sprint or evolves into something more competitive, more dynamic, and, eventually, more balanced.
Advisory services offered through Sowell Management, a Registered Investment Advisor. The views expressed represent the opinion of Sowell Management. The views are subject to change and are not intended as a forecast or guarantee of future results. This material is for informational purposes only. It does not constitute investment advice and is not intended as an endorsement of any specific investment. Stated information is derived from proprietary and non-proprietary sources that have not been independently verified for accuracy or completeness. While Sowell Management believes the information to be accurate and reliable, we do not claim or have responsibility for its completeness, accuracy, or reliability. Statements of future expectations, estimates, projections, and other forward-looking statements are based on available information and Sowell Management’s view as of the time of these statements. Accordingly, such statements are inherently speculative as they are based on assumptions that may involve known and unknown risks and uncertainties. Actual results, performance or events may differ materially from those expressed or implied in such statements. Investing in securities involves risks, including the potential loss of principal. While equities may offer the potential for greater long-term growth than most debt securities, they generally have higher volatility. International investments may involve risk of capital loss from unfavorable fluctuation in currency values, from differences in generally accepted accounting principles, or from economic or political instability in other nations. Past performance is not indicative of future results.
You should receive a confirmation email shortly. Don’t forget to reserve your room through the unique Sowell hotel reservation page. You can extend your stay at the Sowell Summit reduced room rate.
WEEK AHEAD
December 15-19, 2025
A Moment of Market Discord – Hawkish Fed and AI Hangover
Well, the second week of December was truly a class in market contradiction, leaving both stock and bond investors scratching their heads. The headliner was a perfectly executed maneuver by the Federal Reserve, which, with its characteristic professional wit, delivered what can only be described as a "Hawkish Cut."
The Dual Mandate & The Bond Market Enigma
The week’s primary event was the Federal Open Market Committee (FOMC) meeting, where the Fed delivered the anticipated 25-basis-point cut to the policy rate. This move was justified by Chairman Powell's assessment: "Conditions in the labor market appear to be gradually cooling, and inflation remains somewhat elevated." The statement perfectly sums up the core struggle—a dual mandate where the employment side is showing weakness while inflation stubbornly persists.
The bond market's reaction, however, was counterintuitive to a rate cut. Treasury yields rose overall, pushing the prices of existing bonds down. This surprising move was due to nuances in the Fed's outlook and perceptions of underlying economic strength. The yield curve steepened as short-term yields were anchored by the cut, but longer-term yields—driven by inflation fears and the limited future cuts signaled in the "dot plot"—pushed higher. This steepening signaled that, despite the immediate policy easing, the market remained wary of the long-term inflation trajectory.
The Stock Market's AI Cooling
The equity market, represented by the S&P 500 Index, ultimately closed the week down 0.61%. While the immediate post-Fed rally was a flash of bullish exuberance, the weight of a few select sectors proved too heavy.
The culprit? The ever-powerful, but currently temperamental, technology sector. Communication Services and Technology stocks were the primary drag, led by heavyweights like NVIDIA, Google, Broadcom, and Meta. This pressure was the market’s latest demonstration of the familiar boom-and-bust cycle. Any new "shiny object"—from the internet boom and cannabis to crypto, EVs, and now AI—comes with its own journey through exuberance, high expectations, disappointment, and rationalization.
Investors are realizing that even with a transformational technology like AI, not every darling can maintain a trillion-dollar valuation based solely on hype. In a twist of irony, the year-to-date performance champions among the top 10 stocks are actually Google and Broadcom, not NVIDIA, demonstrating the complexity of sector leadership.
An Economic Mixed Bag
The week's economic data only reinforced the ambiguity facing the Fed:
A Nod to the Old Guard and Trade Policy
Amid the volatility of new-age tech stocks, the ultimate safe-haven asset, old-fashioned gold, posted a weekly gain of 2% and substantial YTD appreciation, reportedly up 60% in 2025. This performance underscores investor anxiety, showing a clear appetite for hard assets amidst central bank uncertainty and geopolitical risk.
Finally, the US December Trade deficit of -$52.8 billion continues to narrow, hitting its lowest level since June 2020. This trend, a key policy goal of the current administration, is often attributed to President Trump's tariff policy, which has reshaped global trade flows and domestic production incentives.
Week Ahead
With the drama of the Fed decision and the AI sector's reality check now behind us, the financial spotlight immediately shifts to a barrage of high-impact economic releases next week. If last week was the monetary show, the week of December 15th will be the data deluge.
The market's current narrative—that the Fed is cutting rates but inflation remains a problem—will be severely tested. Key releases include the all-important Consumer Price Index (CPI) on Thursday, which will tell us exactly how stubborn inflation really is, and Tuesday's Unemployment Rate and Nonfarm Payrolls (consolidated for October and November after reporting delays), which will provide the clearest picture yet of the "gradually cooling" labor market Powell described.
Furthermore, the market will digest the ADP Employment Change as a warm-up act and continue to monitor weekly Jobless Claims for confirmation of the cooling trend. Though the PCE (the Fed’s preferred inflation gauge) is scheduled later in the month, the upcoming data provides more than enough fodder to drive bond yields and test the resolve of the S&P 500, ensuring the holiday cheer remains tempered by economic uncertainty.
Fork in the Road in the AI Hardware Race
By Fiona Zhang
Google’s home-grown AI chip, the Tensor Processing Unit (TPU), has quietly graduated from cool internal toy to possible market plot twist. What started as a bespoke accelerator built to run Google’s own AI brainchildren is now popping up in headlines and, more intriguingly, in the tape. And investors are definitely watching.
Lately, whispers that several big AI players are testing or expanding workloads on TPUs have been enough to send traders scrambling. Alphabet catches a bid on those headlines, while Nvidia and AMD see the occasional wobble as the market remembers that the GPU throne might not be welded in place forever. These aren’t seismic rotations yet, more like tremors, but they reveal a market hypersensitive to any hint that AI computing could someday be a multiplayer game.
Part of it is pure capex physics. If even a slice of training or inference workloads shifts toward Google’s cloud-hosted TPUs, that means fewer GPUs purchased elsewhere, and more long-term revenue chugging through Google Cloud. Analysts have already pointed out that faster TPU adoption could brighten Google Cloud’s growth trajectory while casting a bit of shade on pure-play chipmakers.
That’s why even a rumor that a major social platform is “testing TPUs at scale” can vaporize or create billions in market cap in minutes. And you could see that tension around Nvidia’s latest earnings, too: they delivered another beat, yet the stock still finished the day in the red. In this environment, it’s not enough for Nvidia to surprise; they have to surprise the surprise, because competition fear is now part of the trade.
But this isn’t some tidy duel where one chip simply replaces another. GPUs and TPUs aren’t substitutes so much as specialists: GPUs are the general-purpose heavy lifters with massive parallelism; TPUs are laser-focused sprinters designed for the kinds of matrix math that fuels deep learning.
All of this sits atop Nvidia’s deeply entrenched software ecosystem, which remains sticky, beloved, and very hard to replace. No one expects customers to abandon GPUs overnight. Most will run hybrid fleets with GPUs doing some jobs and TPUs doing others, because switching costs are painful and CUDA (Compute Unified Device Architecture) still rules the AI software universe.
Meanwhile, Google still has to prove it can supply, support, and scale TPUs globally across multiple hardware generations — a very non-trivial hurdle.
Chart reference: https://inaccel.medium.com/cpu-gpu-fpga-or-tpu-which-one-to-choose-for-my-machine-learning-training-948902f058e0
Still, markets are telling their own story. For the first time in years, investors can imagine a world where AI compute demand isn’t captured by a single architecture. Even a little diversification in the supply chain could reshape cloud capex, ripple across semiconductor names, and give Alphabet a far bigger footprint in the economics of AI.
For now, TPUs sit in an interesting middle zone: not yet hitting the trend’s peak, but no longer a residual squeak. What they are is a new variable for investors to watch — a signpost indicating whether the AI hardware race remains a one-horse sprint or evolves into something more competitive, more dynamic, and, eventually, more balanced.
Advisory services offered through Sowell Management, a Registered Investment Advisor. The views expressed represent the opinion of Sowell Management. The views are subject to change and are not intended as a forecast or guarantee of future results. This material is for informational purposes only. It does not constitute investment advice and is not intended as an endorsement of any specific investment. Stated information is derived from proprietary and non-proprietary sources that have not been independently verified for accuracy or completeness. While Sowell Management believes the information to be accurate and reliable, we do not claim or have responsibility for its completeness, accuracy, or reliability. Statements of future expectations, estimates, projections, and other forward-looking statements are based on available information and Sowell Management’s view as of the time of these statements. Accordingly, such statements are inherently speculative as they are based on assumptions that may involve known and unknown risks and uncertainties. Actual results, performance or events may differ materially from those expressed or implied in such statements. Investing in securities involves risks, including the potential loss of principal. While equities may offer the potential for greater long-term growth than most debt securities, they generally have higher volatility. International investments may involve risk of capital loss from unfavorable fluctuation in currency values, from differences in generally accepted accounting principles, or from economic or political instability in other nations. Past performance is not indicative of future results.