Intuit dropped 12% on Tuesday. Adobe and ServiceNow each fell 3%. The headlines call it 'AI disruption fears.' I call it a misreading of the signal. The market isn't pricing in the obsolescence of tax software or design tools. It's pricing in a structural shift in how software captures value, and the sell-off has more in common with a margin call than a technological verdict.
Let me be precise. Intuit's core product, TurboTax, is a workflow engine. It guides users through a deterministic series of forms. Adobe's Creative Cloud is a toolkit for creative expression. ServiceNow is a ticketing and workflow platform for IT operations. These are fundamentally different businesses. Yet they all dropped on the same narrative. That's not a coincidence. That's a macro signal.
When three companies with different products, different customer bases, and different revenue models drop simultaneously on a vague 'AI' narrative, the market is telling you something about liquidity, not about technology. The AI narrative is the excuse. The real driver is a rotation out of software with high recurring revenue multiples into anything that looks like pure AI exposure. This is a capital flow phenomenon, not an earnings event.
I've seen this pattern before. In 2022, I tracked the Terra collapse in real-time. The 20% APY loop on Anchor Protocol was not a sustainable yield. It was a liquidity extraction mechanism. When the inflow of new capital slowed, the entire edifice collapsed. The market is doing something similar now. It's extracting liquidity from 'legacy' SaaS names to fund the AI trade. The narrative is just the justification for the capital flow.
Consider the valuation math. Intuit trades at roughly 30x forward earnings. Adobe at 25x. ServiceNow at a staggering 40x. These are not distressed valuations. They are premium multiples assigned to companies with durable competitive advantages. A 12% drop on a narrative is not a repricing of fundamentals. It's a repricing of narrative exposure. The market is saying: 'We don't want to hold software that might be disrupted. We want to hold AI that might disrupt.' That's a momentum trade, not a value trade.
Here's the structural problem. The market is conflating 'AI-native' with 'AI-immune.' It assumes that a company like OpenAI, or a hypothetical AI-native tax filing service, will automatically capture value from Intuit's user base. This is a false equivalence. Distribution matters. Data matters. Trust matters. Intuit has 100 million users who have filed taxes with them for years. That data is a moat. The question is whether Intuit can turn that data into a defensible AI product before a new entrant does.
The core insight is this: The AI disruption thesis is correct in the long run, but the market is pricing it as if it's imminent, which is a liquidity error.
Let me break down the technical reality. Traditional SaaS is built on deterministic logic. If A, then B. The codebase is a series of state machines. AI, on the other hand, is probabilistic. It generates outputs based on statistical inference. These are fundamentally different architectures. Intuit cannot simply add a 'GPT button' to TurboTax and expect it to work. They need to rebuild their data pipeline, their inference infrastructure, and their user interface. This is a multi-year engineering effort.
But here's what the market misses. The cost of inference is dropping exponentially. In 2023, a million tokens cost about $10. In 2025, it's less than $1. This means the marginal cost of delivering AI-powered features is falling. This is good news for incumbents like Intuit, who can deploy AI features at scale without destroying their gross margins. The market is pricing AI as a cost center. In reality, it's a deflationary force that can expand margins if deployed correctly.
My experience with the Compound Finance stress test in 2020 is instructive here. I modeled the protocol's interest rate curves and identified a liquidity crunch risk when collateralization ratios dropped below 150%. The market was focused on TVL growth. I was focused on incentive alignment. The same principle applies here. The market is focused on AI narrative. I'm focused on capital deployment and unit economics.
Let's look at the actual numbers. Intuit has $14 billion in annual revenue. Their R&D budget is roughly $2 billion. If they allocate 20% of that to AI, they're spending $400 million per year on AI development. That's a meaningful investment. But it's also a rounding error compared to the $40 billion in market cap they lost on Tuesday. The market is telling you that Intuit's AI strategy is worth $40 billion less than it was on Monday. That's not a rational assessment. That's a panic.
Now, let me address the contrarian angle. The conventional wisdom is that AI will disrupt SaaS. My thesis is the opposite. AI will entrench the incumbents who have data and distribution, while the new entrants struggle with customer acquisition costs. The data flywheel is the key. Intuit has data on how millions of people file taxes. This data is proprietary. It cannot be scraped from the open web. It's locked inside their systems. This is the kind of data that can train a vertical AI model that is genuinely better than a general-purpose model like GPT-5.
The market is pricing AI as a leveler. I see it as a moat-widener. The companies that have the data to train specialized models will get stronger. The companies that rely on generic AI will get commoditized. This is a subtle but critical distinction. The sell-off in Intuit, Adobe, and ServiceNow is a mispricing of their data assets.
Let me also address the 'decoupling' thesis. The market is treating AI as a separate asset class from traditional software. This is a mistake. AI is a feature, not a business model. The companies that will win are the ones that integrate AI into their existing workflows, not the ones that build AI as a standalone product. OpenAI's ChatGPT is a great consumer product. But it's not a business workflow. Intuit can embed AI into the tax filing workflow and create a seamless experience that no AI-native startup can match.
I've tested this hypothesis in my own portfolio. In January 2024, I executed a basis trading strategy between Bitcoin futures and spot prices, capturing a 2.5% annualized premium. It was a low-risk arbitrage, not a directional bet. The same principle applies to SaaS. The market is offering a mispricing between the 'AI-disrupted' narrative and the actual fundamental value of these companies. The risk-adjusted return on buying Intuit at this level is asymmetric. You're getting a company with a 30x multiple, a dominant market position, and a clear path to AI integration at a discount because of narrative risk.
The takeaway is this: The sell-off is not a signal to exit SaaS. It's a signal to re-enter with a clear understanding of what you're buying.
I'm not saying Intuit, Adobe, and ServiceNow are risk-free. They face genuine challenges. The technical debt of their existing codebases is significant. Their product architecture was not designed for AI. Their go-to-market motion is built around seats, not outcomes. These are real issues. But they are solvable with time and capital. The market is pricing them as if they are existential threats.
Let me close with a forward-looking thought. The next 12-18 months will determine the winners and losers in the AI-SaaS convergence. The incumbents have the data, the distribution, and the customer trust. The new entrants have the technology and the hype. The outcome will be determined by execution, not by narrative. The market's panic on Tuesday was a gift to those who understand the underlying mechanics. Volatility is the tax on unproven consensus. The consensus is that AI will destroy SaaS. The proof is not there yet. The market is paying the tax. I'm collecting the proceeds.