Software must be built to support agents as the primary interface as they increase in volume relative to human users.

“If you have a hundred or a thousand times more agents than people >> then your software has to be built for agents.”

Most companies should consume AI as a managed service rather than attempting to build their own bespoke AI infrastructure.

“one of the most 99% of the companies, I would say what they need is probably to consume this as a managed service in my mind”

The future of SaaS will be defined by agentic commerce where search and browse experiences are collapsed.

“So, today basically your commerce experience is search, browse and then purchase. Search and browse take 95% of time. 5% is then check out and finally everything else, right? Uh So, that search browse experience will move to agentic basically.”

Enterprise data will soon grow at a rate approaching 10 times per year, driven by the increased activity of AI agents.

“imagine what will happen when you'll have 10 million agents running amok on your database, on your customer base, on your data. And so, the amount of activity that is going on right now in the enterprise and is foreseeable in the near future, then you're actually looking at data which is going to go”

Traditional enterprise search is inefficient because it forces employees to navigate between too many disparate applications, each with its own search interface.

“If you're trying to find something at work right now, you're going to have to go to all these individual apps and they have their own search bar. You're going to have 12, 20 different search bars to potentially consider. And that is just kind of ridiculous.”

The modern software market favors consolidation, where providing an all-in-one platform that combines various tools into a single, cost-effective solution creates more value than best-of-breed softwar

“we were on The Cutting Edge of consolidating it and now you see a lot of these tools that were once the best at what they are are now starting to go back saying we're losing we need to now start building these other functions and tools”

Adrian McDermott argues that successful AI adoption in enterprise requires robust orchestration and integration with back-office systems rather than just model improvements.

“So, it's integration and orchestration, integration and orchestration, integration and orchestration, right? In LLM terms, it's all about tool use, right? You have to be able to select the right tool, go get the right information, and follow a procedure.”