When AI Moves Faster Than Your Business Model: Adapt or Perish

 

Anthropic proved software can be built in days, not years. Markets panicked. SaaS cracked. AI won’t kill everyone—but it will expose who’s bluffing, overpriced, and already late to their own funeral.

I watched the screens flicker red and felt the familiar chill run through the market. This time it wasn’t a rate hike, a war headline, or a rogue inflation print. It was code. Not a decade-long rewrite. Not a moonshot product roadmap. Just a few weeks of work, a legal AI tool from Anthropic, and suddenly software stocks were tumbling like dominoes on wet pavement. The sell-off wasn’t polite. It was sharp, fast, and emotional. SaaS and financial services took hits because investors saw something they hadn’t priced in yet: speed that breaks old assumptions.

What shook people wasn’t just that Anthropic launched a legal tool. Legal tech has been around for years. It was the realization that something usable, enterprise-adjacent, and frighteningly capable could be stitched together in roughly 10–14 days. That detail landed like a brick through a window. I could almost hear traders muttering the same question at once: if this can be built that fast, what exactly am I paying for?

That’s when fear takes the wheel. SaaS valuations rest on the idea that software is hard to build, expensive to maintain, and sticky once adopted. Now imagine a world where a lean team with access to a frontier model spins up a product that undercuts years of development. The market didn’t wait to debate nuance. It sold first and asked questions later. I’ve seen this movie before. In the late 1990s, when the internet made distribution cheap, investors declared that brick-and-mortar retail was finished. In the 2000s, when digital ads exploded, print journalism was pronounced dead. Each time, the panic was real, but the extinction story was sloppy.

The legal angle added gasoline. Lawyers are expensive, process-heavy, and allergic to hallucinations. Yet here we are, watching AI draft briefs, summarize discovery, and surface case law in seconds. The irony is brutal. The same profession that bills by the hour is now staring at a machine that doesn’t sleep, doesn’t bill, and doesn’t complain. No wonder markets flinched. But fear has a habit of overshooting reality.

I don’t buy the collapse narrative. I never have. History doesn’t support it. When spreadsheets arrived in the late 1970s, accountants weren’t wiped out. Visicalc and later Excel didn’t destroy the profession; they multiplied its output. By the mid-1990s, Excel was used by over 90% of large firms’ finance teams, and accounting employment kept growing. The same thing happened with databases, cloud computing, and even smartphones. Each wave killed certain tasks, not entire industries. AI is no different. It’s a blade, not a bomb.

What investors are missing is that AI, deployed well, is additive. Chris Kelly hinted at this when he talked about hallucinations and safeguards during his interview at CNBC, a TV channel focusing on business news. Early systems mess up. We’ve already seen hundreds of documented cases where AI generated fake legal citations, invented quotes, and hallucinated precedent. That’s real risk. Lawyers have been sanctioned for filing AI-generated nonsense. But that’s not the end of the story. It’s the beginning of the tuning phase. Every major technology goes through this ugly adolescence. Airplanes crashed before they became safe. Early cars killed pedestrians before traffic laws existed. Nobody concluded that transportation itself was doomed.

The smarter SaaS companies will do what they always do. They will wrap AI in guardrails, checks, audits, and workflows. They will make hallucinations harder, traceability easier, and liability clearer. They will bundle these features, not discount them. People love to argue that cheaper production means cheaper prices. History laughs at that idea. Software pricing has rarely fallen in real terms. What happens instead is feature inflation. You pay the same subscription, but you get more power, more automation, more dependency. The price stays put. The value proposition mutates.

Look at cloud computing. Amazon Web Services launched in 2006 promising cheaper infrastructure. What followed wasn’t a collapse in enterprise IT spending. Global cloud spending crossed $500 billion annually by 2023. Companies didn’t save money. They scaled faster. They spent differently. AI will follow that pattern. Costs may drop at the margin, but demand will explode. When productivity rises, ambition rises with it. A rising tide lifts ships that know how to float.

The market reaction also ignored the partnership dynamic. SaaS firms don’t have to fight AI vendors. They can ally with them. Kelly was right to point out that leaders will partner with companies like Anthropic or OpenAI, embedding models into trusted platforms customers already use. That trust matters. Enterprises don’t want raw models. They want accountability, compliance, privacy controls, and contracts that don’t evaporate when something breaks. That’s where incumbents still have leverage.

Privacy is the quiet killer in this story. AI feeds on data the way fire feeds on oxygen. Upload the wrong document, and you may have just trained a system you don’t control. Governments already know this. There have been documented incidents of sensitive data being used improperly in AI systems, prompting internal bans and policy memos across federal agencies. Enterprises will follow suit. Protocols will harden. Data walls will rise. Companies that can’t manage this will bleed trust, and trust is harder to rebuild than code.

This is where laggards get punished. Not because AI exists, but because they freeze. They deny. They delay. Markets are ruthless with denial. Stocks don’t collapse because technology changes. They collapse because management refuses to adapt. I’ve watched this happen to retailers who ignored e-commerce and media firms who mocked digital ads. The pattern is always the same. First disbelief. Then panic. Then irrelevance.

Anthropic’s legal tool didn’t kill SaaS. It exposed complacency. It forced investors to ask who is actually building, who is bundling, who is securing data, and who is just collecting subscription checks while hoping the storm passes. The sell-off feels overwrought because markets priced apocalypse when the reality is selection. Some firms will get stronger. Others will fade. Capitalism doesn’t do mercy. It does sorting.

I keep coming back to one simple truth. AI doesn’t end work. It raises the bar. It doesn’t erase industries. It strips excuses. The companies that survive this moment will look back and say the same thing winners always say after disruption. We moved fast. We partnered smart. We controlled risk. Everyone else will call it unfair. When the tide goes out, you see who was swimming naked.

 

 

This article stands on its own, but some readers may also enjoy the titles in my  Brief Book Series. You can also read them  here on Google Play: Brief Book Series.