If you have watched the financial news over the past few weeks or if you happen to be an investor in any software stocks, you are likely aware of this sharp and broad based sell-off in these names. The software segment of the market is down over 30% from its September 2025 peak, marking the largest non-recessionary decline in the sector on record. More than $2 trillion (!!) of market cap has been wiped out and its weight in the S&P 500 has fallen from 12% to 8.4% (so far).

What’s behind the decline? Everyone’s favorite two letter word – AI. As announcements were released concerning various large language models (LLMs) and their new/improved functionality (especially in the coding space), the market decided many software products would become obsolete and pushed sentiment off a cliff. However, as an analyst we follow articulates (and we tend to agree), this is about disruption and uncertainty- not about the end (or death) of the software space in its entirety.

Uncertainty Rises, Prices Fall

There is no question that AI will change the world in ways we can’t yet comprehend or imagine. And lately, it seems like the announcements and innovations are coming at us as quickly as the cap ex budgets of the hyperscalers are rising. Investors and markets are not quite sure where things are going to land when all is said and done – and if you have been reading for a while, you know that markets do not thrive on uncertainty. The market is essentially a pricing mechanism of future cash flows and any large unknown variable (like AI) throws those cash flows into question and also raises the rate at which they are present valued (as the risk premium increases). As a result, in the Price/Earnings ratio, the numerator falls, the denominator rises, and the P/E ratio falls dramatically.

Sell First, Analyze Second

This uncertainty has spiked in the software segment as of late, causing investors to sell in mass. There is no doubt that AI will impact portions of the software stack. The challenge right now is finding a way to know which parts will become obsolete, which will remain critical, and which will be enhanced (not replaced) by AI. Right now, the market seems to be fine with “sell now, analyze later” – taking down P/E multiples immediately. This sell-off definitely seems to be more sentiment drive, as opposed to one driven by deteriorating fundamentals in these individual businesses.

Not All is Lost

The arguments against software have been made loudly and consistently in the past few weeks – summarized by the idea that all current software functions can be done by AI for much less money. To date, the most drastic case studies have involved coding, which introduces the ability to rework workflows and potentially shift away from software.

It’s important to remember there are some arguments yet to be made in support of software – including:

*Software services remain embedded in many organizations and come along with deep domain expertise and extensive business integration

*Proprietary customer data sets (contained in these established software programs) is crucial to businesses – and AI does not have the data (yet)

* Many businesses (especially ones that are highly regulated and in complex industries) are slow to change

*Enterprise systems remain highly embedded, have multiple year contracts, and material switching costs – all enhancing software’s moat.

Where We Go From Here

One thing is for certain – this is all just getting started and there will be much more to come on this market-moving topic. It seems likely that some software companies will remain – perhaps with a modify product offering and a varied pricing model. It also seems likely that AI will be embedded into these programs or added to the overall tech stack to work in tandem. So much is uncertain as to how this plays out – and that is what the market is struggling with.

It will be an exciting few years as this all develops! In the meantime, you may just find some attractive investment opportunities in the software space (but proceed with caution and work with your advisor as things are moving very quickly!)

Onward we go,