The 8 AI Anxieties Keeping Leaders Up at Night Heading Into 2027
Most companies heading into 2027 don't have an AI adoption problem. They have an AI results problem, and the pressure to show those results is landing on a handful of overstretched leaders.
That is my takeaway from a sharp piece of field research by Greg Shove, CEO of Section. His team reviewed 600 prospect calls from the past few months and distilled what AI leaders are most anxious about as they plan for next year. It is one of the most honest snapshots of enterprise AI I have seen this year, and full credit for the data goes to Greg and the Section team.
What struck me is how closely their findings match what we hear at SalesXecution, especially inside mid-market and PE-backed companies where the board wants EBITDA, not experiments. Below are the eight anxieties Section identified, and for each one, how we help leaders get out of it.
The eight anxieties, and how we address them
1. “I can't prove ROI”
What Section heard: Leaders can't connect AI spend to business value, and CFOs and boards no longer accept assumed ROI. Even companies with 90%+ daily active use are scrambling as CEOs ask for monthly ROI readouts. As one leader put it, they have adoption, but ROI is still elusive.
How SalesXecution helps: We start with the revenue number, not the tool. Our ARQ diagnostic (AI Revenue Architecture Quotient) baselines where AI can move pipeline, win rates, deal velocity, and cost to serve, so every initiative is tied to a metric the CFO already tracks. Then we put our own money where our mouth is: a portion of our engagement fees is at risk against performance KPIs we set jointly with the client. If the numbers don't move, we don't get paid in full.
2. Adoption is a mile wide and an inch deep
What Section heard: High license and login counts hide trivial chat use. One prospect had 1,500 Copilot users, and nobody was doing more than polishing emails.
How SalesXecution helps: Logins are not a strategy. We embed AI into the actual workflows that produce revenue: account planning, opportunity strategy, call prep, proposal development, and pipeline reviews. Our AI fluency program moves people through defined skill levels tied to their real work, so usage deepens from “rewrite this email” to “build my strategy for this deal.”
3. The board set a mandate with no business context
What Section heard: Top-down pressure is outrunning internal capability. One CEO now requires $6M in net new EBITDA from AI in 2027, with productivity gains excluded, and there is no plan to get there. The sponsor on the call often has no idea how to hit the number they were handed.
How SalesXecution helps: This is exactly where our AI Revenue Architecture Operating System earns its keep. We translate a board target into a value map: which revenue and margin levers AI can realistically pull, how much each is worth, and in what sequence. Having worked with PE-backed portfolio companies, we know how to turn an EBITDA mandate into an operating plan the sponsor can defend in the boardroom.
4. There's no unified strategy
What Section heard: Initiatives are federated or scattered, which leads to duplicated workflows, abandoned agents, and five executives ready to hire five different partners. At one company, the CEO and CFO openly disagree on what AI is for. At another, tooling decisions are frozen until a new hire arrives.
How SalesXecution helps: We align the leadership team before anyone buys anything. Through our Value Architects model, we facilitate a working session that gets the C-suite to agree on what AI is for in their business, which outcomes matter most, and who owns each one. The result is one roadmap, one governance model, and one set of priorities, so the organization stops paying for the same experiment three times.
5. “The army of one”
What Section heard: AI enablement leads are drowning. Section often sees a single person carrying AI enablement for an organization of 10,000+ employees. Those sponsors also carry real career risk from past misfires, rejected proposals, and the exhaustion of being the lone change agent.
How SalesXecution helps: We become that leader's bench. Our team brings deep and varied expertise, including a PhD in change management, a 20-year machine learning and AI architecture veteran, and a 25-year psychotherapist who understands the human side of change. Every consultant is a certified Claude Architect or Claude Developer. We swarm the work so your AI lead can stop doing everything alone and start leading, with wins they can point to.
6. Shadow AI and ungoverned agent sprawl
What Section heard: Employees self-provision tools when they can't get the ones they need. One EVP uses personal accounts because enterprise access is locked down. A healthcare prospect described employees uploading protected health information to unapproved tools.
How SalesXecution helps: Shadow AI is a demand signal, not just a compliance problem. People are telling you what they need. We help you give them sanctioned, secure tools that are actually better than what they found on their own, paired with clear usage policies, guardrails for sensitive data, and an agent inventory so nothing runs without an owner.
7. Cost panic
What Section heard: Consumption pricing has been the biggest enterprise shock of the year, with individual users running up six-figure bills. Approaches to control it are all over the map: hard caps, soft caps with manager approval, centralized model control, or user education on model choice.
How SalesXecution helps: Cost control and value creation have to be designed together. We help you match the right model to the right job, build repeatable workflows instead of expensive one-off prompting, and set spending guardrails tied to business value. When you can see what each dollar of consumption returns, cost conversations get a lot calmer.
8. “Do more with less” headcount mandates riding on AI
What Section heard: Teams are being told headcount will be cut or frozen next year and to figure it out fast, without the enablement or resources to do it. Sometimes the cut already happened and AI is expected to close the gap after the fact.
How SalesXecution helps: We focus on capacity, not just cuts. For revenue teams, that means automating prospecting research, meeting prep, CRM hygiene, and follow-up so sellers spend their time selling. As a HubSpot Gold Partner, we deploy tools like HubSpot's Prospecting Agent alongside an AI platform to give a smaller team the reach of a larger one. And because we teach people how to work with AI, the gains stick after we leave.
The common thread: this isn't a technology problem
Read the eight anxieties together and a pattern jumps out. None of them is really about which model or vendor to pick. They are about strategy, alignment, measurement, and people.
That is why we call our approach AI Revenue Architecture. After three decades in sales performance and transformation, I have watched every technology wave follow the same arc: excitement, scattered pilots, frustration, and finally the companies that win are the ones that redesign how work gets done. AI is the most powerful of those waves by far. It is democratizing expertise in a way we have never seen, and that is a huge opportunity. But capturing it takes architecture, not just access.
Greg closed his post by joking that AI-as-therapist is through the roof because transformation is enough to make anyone crazy. He is not wrong. That is one reason a seasoned psychotherapist sits on our team: real change happens in people before it shows up in the P&L.
Where to start
If any of these eight sounds like your company, start with clarity. Our ARQ diagnostic shows where AI can move your revenue and margin, what is standing in the way, and the sequence that gets you to results your CFO and board will accept.
Schedule a 30-minute conversation with SalesXecution and we will walk through which of these anxieties is costing you the most and what it would take to fix it.
Thanks again to Greg Shove and the team at Section for sharing their research. It is well worth following his work.