There’s a widening gap between what the market says about AI and what we actually hear from customers. The media, the VCs, the AI labs, and influencers have all talked about AI replacing humans, ripping out trusted software, and token-maxxing as ends worth pursuing. But the leaders running real businesses are increasingly asking the right questions. How do I make my people better with AI? Which systems can I trust? How can I measure the ROI of this spend? We hear these questions every day.
After three and a half years of building, shipping, and watching many of our growing customers put AI to work, the AI perspectives we are most certain of at HubSpot are the things almost no one else is saying out loud.
Here are six of them.
AI activity is not AI outcomes.
The industry has confused motion for progress. Drafting emails, generating summaries, doing research. These are activities that AI has made much easier. They are useful capabilities, and we ship them at HubSpot. But activity is the input, not the result. Activity without outcomes is theater.
The companies winning with AI are the ones working backwards from a business problem, not forward from a model demo. For example, customers using Customer Agent are responding to tickets 25% faster, while those using Prospecting Agent are generating 76% more leads.
This is why we moved Customer Agent and Prospecting Agent to outcome-based pricing in April. AI outcomes are what matter. And they’re what we help growing businesses deliver. We put our pricing where our point of view is.
AI is necessary. It is not sufficient.
Generating code is certainly easier now. Anyone can build a prototype in a weekend, but it’s brittle and falls apart under real use. Lowering the floor on generating code doesn’t raise the ceiling on shipping value because the things that actually run a growing business have gotten harder, not easier.
You still need to have clean data, not another silo. You still need to integrate with tens of applications. You still need a full customer view across marketing, sales, and service, one actually powered by context.
The industry will sell you a model or single-purpose agents. But it won’t sell you the system in between: the data hygiene, the workflow design, the change management. That’s left to the customer. And the more disconnected point agents pile up, the harder that work gets.
The future belongs to the companies that build AI into a coherent system, where the data, workflows, agents, and people share context. That’s what we are building at HubSpot. AI is a new layer, not a replacement for the foundation.
AI needs to be built for the Future 5000, not just the Fortune 500.
Today’s AI roadmap is being written for the enterprise that can afford to make it work. By their own disclosures, frontier labs are spending billions of dollars on forward-deployed engineers to get AI running inside large companies.
That model works if you’re a large enterprise. It doesn’t work for the millions of growing businesses that will drive the next decade of growth. A small or midsize company can’t get forward-deployed engineers, rebuild its data pipeline, or build the context platform to make it all work.
So when the consensus says “AI is for everyone,” it’s worth asking who it actually works for today. In practice, it’s the customers who can already afford to make it work, with armies of engineers and developers behind them. That’s not democratization.
We’re optimizing for outcomes per token, not tokens per task.
There’s a business-model conflict in the AI industry that customers haven’t fully seen yet. The vendors who benefit the most from AI usage are not incentivized to make AI cheaper or more efficient. They are incentivized to keep the meter running. So customers are asked to pay for activity and told they are buying transformation.
The honest version of AI economics is the inverse: be clear on the outcome the customer is trying to drive, then find the lowest-cost path to driving it. That is the customer’s job. It should also be the vendor’s. Right now, it isn’t.
Token-maxxing is the vendor’s game. Outcome-maxxing is the customer’s. The vendors that align with the customer will win. The vendors that align with the meter may not.
AI should make people more powerful. Not more replaceable.
The loudest AI narrative is autonomy: agents replace humans, headcount goes down, the future has fewer people in it. That narrative is built for Wall Street, not Main Street. We reject that framing.
We build for the person doing the work, not the person being subtracted from the budget. The rep closing more deals. The marketer shipping more campaigns. The service person solving more complex problems. The owner running more of the business themselves. AI’s job is to make them more powerful, not make them disappear.
Yes, we ship autonomous agents. But autonomy is a capability, not a mandate. Customers decide where to delegate, where to keep humans in the workflow, and where AI suggests. Our defaults are built to serve the operator, not slash the org chart.
We believe in human authenticity and AI efficiency. The things AI cannot replace — trust, judgment, taste, relationship will only get more valuable as the things AI can do become ubiquitous. The companies betting against the human are going to lose the customer, the employee, and eventually the public, of which 57% already think the risks of AI outweigh its benefits.
Trust is more than a privacy policy.
Every AI vendor is claiming trust. But most define it as a security posture: we won’t train on your data, we’re SOC 2 compliant, we offer enterprise SSO. Those things matter. They are also table-stakes. None of them is a differentiated claim. They are what you promise.
What you prove is something else. Real trust is a complete business posture: how you choose the model and handle cost, reliability, and governance for your agents. That’s what customers are actually asking for. Can I trust the model choice? Can I trust the cost? Can I trust the reliability? Can I trust the governance?
Privacy answers what we won’t do. Trust answers what we will. Most of the industry is still answering the first question. The second is the one customers need.
What this all adds up to
The AI consensus held so long as no one in the room had to answer for it. Cut headcount. Rip out the old stack. Keep the meter running. Trust us.
Growing businesses cannot spend time cutting through what is hype versus what is reality. They do not have forward-deployed engineers to throw at implementation. They cannot absorb a pricing model that bills for activity and calls it transformation. They cannot build on a stack that treats humans as the exception.
They need AI built on a foundation that works for them, designed to empower and not eliminate their people, and delivered by a vendor whose business model is aligned with theirs, not against it.
That is what we are building at HubSpot.