The UiPath winds of change and six month investment update after a partial retail investor washout.

As an entrepreneur and business owner, I am sensing more wind in the sails of UiPath, as it’s management acknowledges the fruits of their agentic transition.  Admittedly, the company operates at a corner of the office building far removed from small business and regular entrepreneurs, but it’s an underdog itself in the entrepreneurial world among behemoth companies. It’s steadily growing and adapting its business model within its current trusted relationships (and new ones) with corporate and government clients.

The CEO, Daniel Dines, is also starting a substack and podcast to reach out in a marketing effort.  As with their stock buybacks, can’t imagine they are doing this if they have no good product to offer. 

His latest, “Between Panic and Euphoria”, details the transition from the SaaSpocalypse to zoning in on those who are charting a “path” forward while the media operates in panic job losses/deflation and euphoria of “Abundance” everywhere.

Perhaps like the discovery of electricity, there is a middle ground.  Even the CEO of JPM, Jamie Dimon, said to Andrew Ross Sorkin that there will be “lots of token management” with LLMs and that companies will start going to the cheapest provider.  He said that JPM will do everything it takes to protect its data, and he feels it’s in the “earliest stages”.  “Of course, they are going to be rational about it just like any other resource they use.”  Even Chamath Palihapitiya said he spoke to his CFO, and they are approaching the asymptote on productivity/ROI gains from their token spend.  On the euphoric end, you have the Moonshot guys almost saying don’t save any money, because there will be “Abundance” soon.  Sounds awfully like the “Technocracy” movement of the 1920s where Technologist leaders said the same thing.

Now comes Dine’s Substack: “Between Panic and Euphoria, The Operating Model”…. “I keep hearing some version of the same hope: the next model release will automate the business. Better models will matter enormously, but intelligence is only one layer. The company still has to make the work legible by defining its objects, states, rules, permissions, gates, owners, and execution rails. That is what allows AI to contribute inside a system where people retain ownership of judgment and consequence.  This changes where AI transformation should begin. Start with one meaningful workflow. Map the work, assign the rights, define the boundary between proposal, decision, and execution, and preserve the human knowledge the workflow has quietly depended on.”

LLMs won’t magically turn every company into an autonomous profit engine with a vibe code.  Raw model intelligence is necessary but needs an operating model like UiPath’s that makes the work legible and safe to both humans and robots.  The panic (“AI will replace us all”) and euphoria (“we’re almost there with another prompt of potential superintelligence”) miss the mark wildly.  Intelligence without structure can be expensive noise.  Most companies run on tribal knowledge, implicit handoffs, and particular individuals who “just know” how things work. AI can’t reliably act inside that fog. It needs objects and core entities such customer records, deals, campaigns, support tickets, etc., to work on.

States (what are the possible statuses and transitions?)

Rules & permissions (who can change what, under which conditions?)

Gates (approval points, risk thresholds, compliance checks)

Owners & rails (accountability + execution constraints)

This is essentially turning the disparate corporate reality into a domain model that an AI agent (or ensemble of agents) can reason over, propose actions in, and execute within bounds. Without it, you get potential hallucinated steps, permission violations, or “AI did something weird again” moments that erode trust.  Where you actually start a ROI process at a company is by one meaningful and tested workflow.  It won’t be a company-wide AI transformation in an ocean of AI agents everywhere with no boundaries.

They’ll map new workflows with potential ROI ruthlessly and document every step, decision, and data flow just as Jamie Dimon said.  Human judgment will stay as with ethics, customer empathy and exceptions.

So as an entrepreneur myself, the underdog UiPath will be a fun long-term investment to watch.

The electrification of factories (roughly 1880s–1920s) is one of the strongest historical metaphors here  with hype and panic.   Thomas Edison’s light bulb and central power stations sparked massive excitement in the late 19th century. Electricity promised cleaner, more flexible power than steam engines, water wheels and gas. Investors poured in, cities lit up, and industrial revolutionary productivity was possible.  There was also panic: fears of dangerous overhead wires killing people, job losses and potentially unreliable new technology.  There were stories of people putting buckets under their outlets to catch unused electric going onto the floor.  Yes, the system was gradually redesigned – especially going from old gas lighting in homes to rewiring knob and tube.  Real productivity exploded in the 1910s–1920s, once the complementary innovations (e.g., reorganized factories, new management approaches) caught up.  The power (intelligence) was available early, but the organizational and physical operating model took decades to build. Yes, those who clung to the old systems such as steam and gas lighting got crushed, but those who invested in the rails and boundaries won big.

So in conclusion, I am about even on my UiPath share purchases, but I am not selling.  It still seems that there is so much potential here, and the winds of change are occurring out of the ashes of the SaaSpocalypse.  A lot of very public retail investors have been washed out, but the long-term cool hands may be in charge now.  I would imagine the next milestone is when the shorts have their Waterloo, and it is 100% clear that PATH is not going to be vibe coded into non-existence  I wonder what price that will be.  From thereon, analyzing the new narrative and numbers will be the tricky part, but I’d welcome that problem vs the one now. Plus, the CEO should get some credit for being right on the ball the last year with the cool-headed orchestration narrative vs the vibe code anything to apocalypse one.

Not selling my shares and looking forward to “orchestrating” a narrative reset later this year.  (Not Investment Advice.)