From a Productivity Tool to a Market Disruptor
How AI Is Forcing a Re-Rating of Software, Finance, and Law
By: Brandon Chi
Feb. 11, 2026
For most of the past three years, artificial intelligence has been framed as a productivity accelerant. A tool that makes workers faster, software stickier, and overall produce better margins. We saw the replacement of a lot of service-related roles being replaced with artificial intelligence, with mass layoffs happening these past years. However, the question remained for investors and graduates of finance, and law of how these tools will shape the industry. For investors, the narrative is starting to crack. Equity markets largely rewarded companies that could slap “AI-enabled” onto a slide deck or an earnings call and promise incremental efficiency gains.
Recent market reactions: sparked in part by advances from Anthropic and its large language model Claude, and enterprise platforms like Harvey, suggest investors are no longer viewing AI as just a helper. They’re beginning to price these securities as a substitute, and the equity market is responding accordingly.
Why Software Was Re-Rated First
Software was always going to be the first sector to feel the pressure after the service industry. Not due to growth suddenly disappearing; Y Combinator had around 63% of their W25 cohort being SaaS, with 58% of the entire batch focused on AI-powered B2B. However, defensibility is coming into question.
Most SaaS businesses are built on a familiar formula: per-seat licensing, recurring subscriptions, churn rates, and embedded workflows that justify the premium pricing. Claude-style AI agents challenge this structure by collapsing entire workflows into a single interface. Instead of logging into multiple platforms, such as Salesforce (NYSE: CRM), users can now ask one model to analyze documents, summarize data & meetings, generate insights, and draft these outputs all in real time.
This change matters at the margin, and markets are margin driven. When the same output can be generated faster, and more importantly, cheaper, pricing power becomes harder to defend. This is why analytics platforms, compliance software, and workflow-heavy SaaS names have faced disproportionate pressure. The concern is not that these products will disappear overnight, but that AI will gradually absorb enough of the underlying function to compress these margins, and slow long-term revenue growth.
Equity markets are forward-looking; a huge reason why we saw a lot of disruptions in 2025. Even the expectation of substitution is enough to force a sell-off.
Finance: Claude, Excel, and Financial Modeling
For years, finance was viewed as more insulated. AI struggled with structed reasoning, and early attempts at financial modeling were unimpressive. Wall Street tested AI-generated DCFs, sensitivity and scenario analysis, and valuation frameworks that were dismissed as unreliable. The logic and investment thesis’s were shaky, inconsistent, and the output required more fixing than saved time.
However, this may no longer be the case.
With recent improvements in reasoning-focused models like Claude, and their integration into familiar tools such as Microsoft Excel through enterprise AI layers, financial modeling has quietly crossed a threshold. Claude can now construct full DCF frameworks, layer in bull, base, and bear cases, and flex assumptions dynamically. Not only this, but reason the drivers behind valuation.
Although the street does not believe it can replace junior analysts, it does compress workflow. What once took hours of manual modeling, and error checking, can now be drafted in minutes, and can help analysts with better assumptions, risk assessment, and interpretation rather than construction of these models. The question is now how financial institutions can implement this, without the risk of data breaches and their analysts putting their balance sheets into these language models.
Law and the First True Labor Shock
If software was re-rated on pricing power and finance on workflow compression, law represents something even deeper: a direct challenge to labor.
Legal AI Startup Harvey in Talks to Raise $200 Million at $11 Billion Valuation, increasing its valuation from $8 billion in December.
Legal services have historically relied on time-based billing, junior associates, and information. Harvey directly targets that model. Purpose-built for legal reasoning, document analysis, and drafting; it automates work that once justified hours of billable time. Research that took days can now be completed in hours and minutes, and first-pass drafting no longer may require a full team.
However, this does not eliminate lawyers. But it will reduce the economic leverage embedded in the traditional business model. For investors, the distinction between is critical. Legal data providers, research platforms, and legal software are now being evaluated through a different lens. Markets are not waiting for courtrooms to change but discounting the shift now. It will be interesting to see where artificial intelligence can be seen through a lens most people haven’t seen.
The bottom line is that Claude did not break software or finance, or Harvey did not break law, but together, they forced markets to confront a harder truth for this year. That AI is no longer a tool inside the workflow, but becoming the workflow.
