Somewhere in a conference room right now, a business executive is watching a demo of an AI agent. It books a meeting, pulls a report, fires off a Slack message, and updates a CRM record — all without a single human click. The room goes quiet. Then someone asks the question nobody wants to answer out loud: so what do we do with all the people who used to do that?
That tension sits right at the center of the AI agents conversation in 2026, and it’s why an interview like this one with Darin Patterson, VP of Market Strategy at Make, cuts through the noise. Patterson isn’t selling fear or hype. He’s selling something rarer: a framework for actually making AI agents work inside real organizations, with real humans, at real scale.
The facts:
- Darin Patterson is Vice President of Market Strategy at Make, a platform built specifically for AI agents and automation.
- Make’s stated philosophy is that companies should “invest in people as much as, if not more than, they invest in AI.”
- AI agents can independently search databases, compare data, prepare responses, update systems, send notifications, and monitor tasks to completion — without human prompting at each step.
- By early 2025, nearly every major software vendor had either embedded generative AI in its products or enabled it as an add-on, according to reporting by NoJitter.
- Agentic AI — agents that act on objectives rather than commands — is now being compared in significance to the arrival of cloud computing.
The Gap Between “We’re Using AI” and Actually Using AI
Most companies that say they’re deploying AI are really running experiments. Isolated pilots. A chatbot here, a summarization tool there. Patterson calls this the experimentation trap, and it’s where the majority of enterprise AI investment currently lives and dies. The problem isn’t the technology. The problem is infrastructure — specifically, the absence of process visibility that would let AI agents operate reliably in production environments rather than controlled demos.
An AI agent is only as useful as the systems it can actually touch. An ordinary generative AI system answers a question or drafts a document. An AI agent goes further — it can search a database, compare results, update a downstream platform, and loop back to verify completion. That’s not a chatbot. That’s a digital worker. And digital workers need organizational scaffolding the same way human workers do: clear processes, defined handoffs, and someone accountable when something breaks.
Patterson’s argument — and it’s a sound one — is that conversational AI has fundamentally changed who gets to participate in automation. You no longer need to be a developer to build an automated workflow. That’s real. But the companies getting actual value from it are the ones who paired AI capability with process discipline. They mapped their workflows before they automated them. They know what the agent is doing and why. Everyone else is just generating impressive screenshots for board decks.
This matters because the failure mode here is expensive. A company that automates a broken process doesn’t fix the process — it breaks it faster, at scale, with no human catching the errors. Security and data governance risks in agentic AI environments are real and underreported. When an agent can autonomously access multiple connected systems, the attack surface for a bad actor — or a poorly configured permissions model — grows significantly. Patterson’s people-first framing isn’t just feel-good messaging. It’s also quietly a risk management strategy.
Is the “Humans Plus AI” Framing Just Corporate Cover?
Here’s the honest take: the industry’s current obsession with “AI as a teammate, not a replacement” is partly sincere and partly self-serving. Amazon Connect frames its AI model around hybrid collaboration. Make frames it around human investment. Every major vendor is racing to package the same thing — agents that work alongside people — because the alternative framing (agents that replace people) is a PR and regulatory nightmare.
But there’s a real idea underneath the soft language. The companies that are genuinely winning with AI agents right now are not the ones who ripped out human roles and plugged in software. They’re the ones who redeployed human attention. The accountant isn’t preparing routine reports anymore. The customer service rep isn’t answering the same four questions for the hundredth time. The manager isn’t pulling dashboard numbers manually. The agent handles the rote. The human handles the judgment. That division of labor, done well, is genuinely powerful.
The question that nobody in enterprise tech wants to answer honestly is what happens to the workers who are the rote. Patterson talks about investing in people. That’s admirable as a principle. As a workforce policy, it needs a lot more specificity than a keynote can provide. Watching how companies navigate the gap between that principle and their actual headcount decisions over the next 18 months will tell you everything about whether “people-first AI” was a philosophy or a talking point.
The AI agents story shares something with other big infrastructure bets — the same “what happens to the humans” question follows every major efficiency breakthrough, whether it’s a six-minute EV recharge displacing gas station workers or low-cost water purification disrupting existing infrastructure labor markets. Efficiency is politically neutral until it hits a paycheck.
The companies that figure out process visibility first — before they scale agents — will set the standards everyone else copies, and the gap between them and the slow movers is already starting to widen in ways that are going to be very hard to close.
Watch the Breakdown
Sources
- Q&A: Make’s VP at Darin Patterson on AI Strategy — aimagazine.com
- 5 ways AI shifts organizational workflows — www.nojitter.com
- AI Agents Are Changing Information Technology: Are Businesses Ready for Software That Can Act on Its Own? — www.modernghana.com
