AI Tasks for Teams: The Point Is to Be More Human, Not Less
A sales rep I was working with last month pulled up a customer email he'd drafted with AI. He read it out loud to me. Halfway through, he stopped and said, "This doesn't even sound like me." He was right. It was polished, grammatically perfect, and completely forgettable. His client would have read it, shrugged, likely rolled their eyes, and moved on with their day.
Funny thing is - that same rep was spending three hours a week manually updating his pipeline notes after every call. Copying details from one system to another. Reformatting a forecast spreadsheet his manager wanted by Friday. All things taking him away from his customers.
There's a lot of conversation right now about AI tasks for teams, especially in sales. Most of it focuses on using AI to write outreach emails, draft proposals, or generate client-facing content. And I get the appeal. Those tasks take time. But when I look at how teams actually spend their days, the biggest time drains aren't the customer conversations. They're the invisible tasks that pile up around them. The CRM updates. The meetings people sit in and multi-task. The approval chains that sit in someone's inbox for two days.
That's where AI should live. The stuff that gets in the way of the good stuff.
AI Tasks vs. Human Tasks
I use a framework called BORE when I help teams figure out where to start with AI. It stands for Boring, Overhead, Relationship-less, and Easy to delegate. If a task checks one or more of those boxes, AI is probably a better fit for it than you are.
That "R" is the one I find myself talking about the most lately. Ask yourself: does anyone need to feel something from this task? A thank-you note to a client, a tough check-in with a teammate, a congratulations on a win? Those need a human. A first-draft process doc? That's relationship-less. Hand it over.
Think about your average week. How much of it is spent on work that doesn't require your expertise, your judgment, or your relationships? For most teams I work with, the answer is somewhere between 30% and 50%. That's what shows up when we actually map out AI tasks for teams and compare them to what people are doing manually.
Here's a reference table of AI tasks for teams versus the ones that should stay human.
The left column is where AI should do the heavy lifting. The right column is where you stay in the driver's seat (with a few notes on how AI can still play a supporting role).

The best AI tasks for teams are the ones that are repetitive, structured, and relationship-less. No human moment required. The moment a task depends on trust, nuance, tone, or connection, that's your lane.
Using AI as a thought partner (not a ghostwriter)
One thing I want to be careful about: I'm not saying AI has no role in communication tasks. It does. But the role is support, not substitution.
Before a big client meeting, ask AI to pull together a briefing on the account, including recent activity, open issues, and key contacts.
Before writing a proposal, ask AI to summarize what the client said they cared about across three different calls.
Before a tough conversation with a direct report, ask AI to help you organize your talking points and think through how different approaches might land.
That's AI as a thought partner. You're still the one showing up, making decisions, and bringing the empathy that no model can replicate. AI just makes sure you're walking in prepared instead of winging it. It gives you the structure and confidence so you can focus on the good stuff.
This maps to what I call the Workflow Evolution path. The goal isn't to jump straight from "I do everything" to "AI handles it." For most of the human-centered tasks in that right column, the sweet spot is "AI helps me do it." You keep the wheel. AI rides shotgun with the map open.
Try This: The 30-Minute Task Sort
Block 30 minutes with your team this week. Have everyone write down their top 10 recurring weekly tasks on sticky notes (or in a shared doc). Then sort them into two buckets:
BORE tasks (Boring, Overhead, Relationship-less, Easy to delegate): These are your AI tasks for teams. They go in the "hand it to AI" pile.
Human tasks (require judgment, empathy, creativity, or relationship): These stay with you, but write one way AI could support each one.
You'll probably agree on 70% of the list immediately. The 30% you debate? That's the interesting conversation. That's where your team starts building real clarity about how AI fits into their work.
The goal of AI was never to make your team sound like a chatbot. It was to give your people back the hours they're losing to busywork so they can do more of the work that actually matters. More prep before the big call. More time thinking through a tricky negotiation. More energy for the conversation that builds a relationship instead of just closing a deal.
Keep human stuff, human.
Chevelle



The contrast between the polished email that “doesn’t sound like me” and three hours spent updating pipeline notes really nails it: automate the admin, not the humanity. I’ve been using https://art-list.pro/art-styles/anime
The “polished but completely forgettable” email nails the real issue: AI should remove pipeline busywork so reps can sound more like themselves. I’ve been using https://soroai.pro
The “polished, grammatically perfect, and completely forgettable” email nails the risk of using AI without your voice. Automating pipeline notes seems like a better way to make room for real conversations. I’ve been using https://revidai.pro