Tech shares are getting hammered, and the weakness is spreading across the market. The common thread is artificial intelligence. Investors are rethinking the cost, the politics, and the payoff of the AI buildout. I’m Taylor Sohns, CEO of LifeGoal Wealth Advisors, a CIMA and CFP. I’ve spent years studying market cycles and investor behavior. Today I’m laying out why this selloff is happening, what matters most, and what I’m watching next.
What Sparked the Latest Selloff
Three pressure points are rattling confidence: leadership tone, political pushback, and doubts about demand. Each one touches a different piece of the AI story—capital, power, and profits. Together, they explain the sharp move lower in tech and why the pain spread to the broader indexes.
“Tech stocks are being taken to the woodshed, causing the entire market to get smoked.”
That reaction isn’t coming out of nowhere. AI still needs massive amounts of money, huge energy, and clear revenue. When any one of those looks shaky, the market flinches. When all three look shaky at once, it sells first and asks questions later.
View this post on Instagram
The Flashpoint: Sam Altman’s Exchange With Brad Gerstner
OpenAI’s spending plans became a lightning rod. Investor Brad Gerstner pressed Sam Altman on how a company with about $13 billion in revenue can commit to a spending number as high as $1.4 trillion over time. That’s a 100x-plus gap. It deserves a straight answer about financing, timing, and partnerships.
“Brad Gerstner asked him how OpenAI, with a $13,000,000,000 revenue, can commit to spending 1,400,000,000,000.0. Instead of addressing the legit question, he says, Hey, Brad, if you wanna sell your shares, I’ll find a buyer.”
The response set off a firestorm. Some called Altman a visionary. Others saw it as a dismissive dodge. Markets crave clarity when the numbers are that large. The tone of leadership, fair or not, becomes part of the risk premium investors assign to a stock or a theme.
Here’s why the number matters. Building and running AI at scale takes data centers, chips, power, networking gear, land, water, and staff. The price tag runs into the hundreds of billions across the industry. If leaders frame those costs without a clear plan for funding and payback, investors ask whether the math is credible. That uncertainty alone can clip valuations.
Does the Financing Math Add Up?
Let’s be blunt. A trillion-dollar spending path demands a detailed map. It would likely require:
- Partnerships and joint ventures: Cloud providers, chipmakers, and utilities sharing costs and profits.
- Staged investment cycles: Phasing spend based on demand thresholds and utilization targets.
- Blend of equity and debt: Spreading risk across capital sources.
- Prepaid contracts: Long-term customer deals that secure cash flows and improve financing terms.
Without those pieces, the burden falls on balance sheets and future capital raises. Markets won’t accept that at face value. They want proof: signed customers, high utilization, and clear returns on each build phase. Until then, price-to-sales and price-to-earnings multiples compress. That’s what we’re seeing.
Political and Social Pushback on Data Centers
The second pressure point sits in local communities. Many voters are pushing back on new data centers. They worry about higher power bills, water use, noise, and land use. Those concerns are not trivial. AI compute is power-hungry, and large sites can pull as much electricity as small cities.
“Voters are voting against data center build outs in their community out of fear of electricity prices skyrocketing. AI doesn’t exist without the data center.”
Permitting risks slow projects, raise costs, or shift them to less suitable locations. Even if a project gets approved, utilities need to build transmission lines and substations. That takes years. If energy supply struggles to keep up, operating costs rise. That squeezes margins and delays returns.
Many tech leaders pitch cleaner energy to solve this. That’s helpful, but it doesn’t cut through near-term limits. Wind and solar need storage to match data center loads. Nuclear takes time and faces public resistance. Gas remains a bridge, but it raises emissions debates and policy risks. The outcome is more uncertainty and more cost.
Will AI Demand Meet the Hype?
The third pressure point is the one that matters most to stock prices: revenues. The market is asking whether the explosive demand forecasts will show up on time and at profitable prices.
“Huge questions around whether the insane demand forecast from the big AI players will ever actually materialize. Today, the market’s saying, hell no.”
To justify massive spending, AI needs repeatable, high-value use cases. Some are here: coding assist, call center routing, content drafting, fraud detection, and ad targeting. These save time and boost output. But they must be priced and sold in a way that covers compute costs and capital charges with room to spare.
Two questions keep coming up:
1) Utilization: Are those expensive graphics chips and data halls running at high, steady usage, or do they sit idle after initial experiments?
2) Monetization: Are customers paying premium prices for finished tools, or are they stuck in trials and pilots with low revenue per user?
If utilization stays choppy and pricing stays soft, the return on invested capital disappoints. That forces spending cuts, delays new builds, and hits growth narratives. Stocks reprice fast when that happens.
What the Market Is Pricing Right Now
In my view, investors are discounting three near-term risks:
- Financing friction: Tougher capital markets and pushback on vague spending plans.
- Energy and permits: Longer timelines and higher operating costs for mega sites.
- Revenue lag: A slower shift from demos to paid, scalable products.
That discount shows up in falling multiples, even for leaders. It also spills over to suppliers—chipmakers, power equipment firms, and real estate owners tied to data centers. When the biggest buyers pause or message caution, everyone in the chain feels it.
Signals That Would Restore Confidence
To turn sentiment, investors will look for concrete proof points. Here are the markers I’m tracking:
1) Contracted demand: Multi-year, prepaid deals with large enterprises that tie spend to usage milestones. Prepayment changes the cash flow math and lowers capital risk.
2) Utilization data: Clear metrics on GPU hours, inference vs. training mix, and sustained usage after initial rollouts.
3) Product stickiness: Higher renewal rates and wallet share as AI tools move from testing to daily workflow.
4) Energy plans with timelines: Signed power purchase agreements, new transmission projects, and on-site solutions that match demand curves.
5) Phased capex discipline: Spend that steps up only when KPIs are met, with returns tied to each build stage.
Markets don’t need perfection. They need visibility. A few quarters of clean execution could change the tone fast.
How I’m Thinking About Positioning
I’m not writing off AI. I am respecting the cost and timing. In my work with clients, balance and time horizon come first. Growth themes can fit, but size them with care and check concentration risk. Many portfolios drifted heavily into mega-cap tech. That added return on the way up. It adds volatility on the way down.
For investors with long time horizons, dollar-cost averaging can help reduce the urge to time the market. For those with shorter timelines, risk control matters more than catching the bottom. I also look at second-order plays with steadier cash flows—utilities improving grid capacity, select equipment makers with booked orders, and software firms that package AI into specific, paid outcomes.
Key Takeaways
- Leadership tone matters when the spending numbers are huge; clear plans beat bravado.
- Permitting and power are real bottlenecks; delays and costs affect returns.
- Revenue proof will decide the winners; utilization and pricing need to hold up.
- Markets are punishing vague stories and rewarding execution with measurable metrics.
- Position size and time horizon should match your tolerance for sharp drawdowns.
A Final Word
AI is not going away, but the market just reminded everyone that money, power, and profits must line up. The recent selloff is a check on narratives that got ahead of the math. As an investor and advisor, I focus on signals that turn stories into cash flows: contracted demand, sustained usage, and disciplined spending. Those are the signs that will separate durable growth from hype. Until they show up, expect volatility—and use it with care.
