When AI answers, someone stops feeling needed
Every question that goes to Copilot instead of a colleague removes a small signal that the colleague matters. HBR's Zach Mercurio calls that feeling mattering, and he links it to engagement and performance.
Mercurio opens his November 12, 2026 article in Harvard Business Review with a manager at a building materials company who told him, "No one really asks me anything anymore. They just ask Copilot. Sometimes I wonder, 'What's my point?'" The article's abstract defines mattering as the experiences that help people feel seen, affirmed and needed. It links that feeling to engagement, commitment, and individual and organizational performance, and it says leaders need to take steps to protect it.
The habit behind that manager's comment is not hostile. In Microsoft's 2025 Work Trend Index, the top reasons people gave for turning to AI instead of a colleague or manager were 24/7 availability (42%), machine speed and quality (30%), and unlimited ideas on demand (28%). Reasons about avoiding colleagues, such as their impatience or judgment, ranked lowest. People are not trying to cut each other out. The chatbot is simply always there.
That chatbot is already on most teams. As we found in our look at what AI tools small and mid-sized businesses actually use, most privately held businesses run on a general-purpose chatbot, usually ChatGPT, with Microsoft Copilot common wherever a company already works in Microsoft 365. The problem is not the tool. It is what quietly disappears when nobody decides which questions still belong to people.
Remote teams feel it first
On a remote team, the quick chat question was often the main way people connected, so remote employees lose the most when those questions go to AI instead.
Gallup's global data shows why that matters. In its State of the Global Workplace: 2026 report, which covers global 2025 data, exclusively remote employees were the most engaged work-location group, at 30%. They were also tied with hybrid employees for the highest daily loneliness, at 24%. In other words, remote employees can be engaged with their work and still lonely.
More recent U.S. data points the wrong way. In Gallup's October 28, 2026 report, Jim Harter reported that U.S. engagement fell to 29% in the third quarter of 2026. Engagement among exclusively remote employees fell five points, from 37% to 32%. Among exclusively remote employees at organizations that have integrated AI, it fell from 38% to 33%. Gallup also found employees were less likely to feel that someone at work cares about them as a person. Gallup does not say AI caused the drop. It says the findings raise questions about whether communication and manager support are keeping pace with AI integration.
Chat is also where AI work now starts. As we covered when GitHub Copilot in Slack and Teams became generally available, a message in a team channel can now kick off an AI coding session. For teams that turn it on, the channel where people used to ask each other things can become the place where they ask the machine.
Write a short "Ask AI / Ask a person" note
A one-page team note on which questions go to AI and which go to a person keeps human contact deliberate instead of accidental. Mercurio's first recommendation, in the article's abstract, is to establish "Ask AI / Ask a person" norms and preserve meaningful human interactions.
Clarity pays off. In Gallup's July 21, 2026 analysis, Harter found that among employees at AI-adopting organizations, 43% were engaged when their organization had a clear plan for integrating AI, versus 28% without one. Manager support mattered even more: 48% of employees who strongly agreed their manager actively supports the team's AI use were engaged, versus 30% of those with lower agreement. Those figures measure an organization's AI plan and a manager's support, not a team note. But the lesson carries over: people engage more when someone tells them how AI fits their work.
A useful note is short and specific. For example:
- Ask AI: formatting, first drafts, summarizing a long document, syntax and spreadsheet formulas, rewording an email.
- Ask a person: client history, why a past decision was made, judgment calls, anything a new hire is learning in their first months, and anything where a colleague's experience is the point.
- Name names: list who to ask for what, so "ask a person" leads to a real person.
Review the note every quarter. The point is not to ration AI. It is to make sure the questions that build trust between people still get asked.
Make human contributions visible
When AI drafts and people review, managers should name the judgment each person added, so that review work does not become invisible. Mercurio's abstract lists making human contributions visible among the steps leaders should take.
Review is now a normal part of working with AI. In Microsoft's 2026 Work Trend Index, 86% of AI users surveyed said they treat AI output as a starting point, not a final answer, and that they stay responsible for the thinking. That thinking is where people add value, and it is also the easiest part of the job to overlook. A polished deliverable looks the same whether someone caught three errors in it or none.
Managers can fix that with small habits:
- In team updates, say what a person changed or caught, not just that the work shipped.
- In shared documents, note who reviewed AI-drafted work and what they decided.
- In one-on-ones, ask about the calls a person made, not only the output they delivered.
Naming the judgment tells people their part was seen, and seen is the first part of mattering.
Protect idea generation
Let people originate ideas and then develop them together with AI, rather than having each person prompt alone. Mercurio's abstract names protecting opportunities to originate ideas as one of the steps leaders should take.
The research favors people and AI working together. HBS Working Knowledge reported on October 17, 2025 on a field experiment with 791 Procter & Gamble professionals. Ideas ranking in the top 10% were three times more likely to come from teams using AI than from individuals working without AI. As coauthor Fabrizio Dell'Acqua put it, "If you want to be in that top 10% of performers, a full human team plus AI seems like the recipe for success."
For a remote team, that suggests a simple sequence:
- Ask each person to bring their own ideas before anyone opens a chatbot.
- Share and discuss those ideas on a call or in a shared document.
- Use AI together, in the same session, to extend, combine and stress-test what the group produced.
Reinvest AI time savings in people
Spend the hours AI frees up on judgment, empathy and personal support, and tell your team that this is why they remain needed. Mercurio's abstract says leaders should clarify why people remain needed and reinvest AI time savings in uniquely human contributions such as empathy, judgment and personalized support.
Freed time does not go to people on its own; someone has to decide where it goes. For a manager of a remote team, deliberate choices might include more one-on-one time, mentoring newer staff, spending longer with clients on hard decisions, or giving a teammate space to work on something that uses their strengths.
Then say it out loud. If AI now handles the first draft of a report, tell the person who used to write it what you want from them instead, and why it matters. People who know why they are needed have an answer to the question that manager asked: what's my point?
We will look at who decides where that freed time goes in an upcoming article, "AI gave your team back a third of the time on the tasks it took. Who decided where it goes?"
We can help you with that.
NTx AI Consulting works with owner-led and partner-led privately held businesses that are currently using AI tools in everyday workflows, or would like to be. The complimentary consultation is a working session, not a product demo. If a deeper engagement makes sense, Phase 1 is an AI Disruption Readiness Audit. You can learn more at https://ntx.ventures/ai-consulting/.
Sources
- his November 12, 2026 article in Harvard Business Review. https://hbr.org/2026/11/how-ai-erodes-peoples-sense-of-value-at-work
- article's abstract. https://store.hbr.org/product/how-ai-erodes-people-s-sense-of-value-at-work/H09DHG
- Microsoft's 2025 Work Trend Index. https://www.microsoft.com/en-us/worklab/work-trend-index/2025-the-year-the-frontier-firm-is-born
- State of the Global Workplace: 2026 report. https://www.gallup.com/workplace/697904/state-of-the-global-workplace-global-data.aspx
- Gallup's October 28, 2026 report. https://www.gallup.com/workplace/716117/employee-engagement-below-30-first-time-years.aspx
- Gallup's July 21, 2026 analysis. https://www.gallup.com/workplace/712433/employee-engagement-remains-flat-adoption-accelerates.aspx
- Microsoft's 2026 Work Trend Index. https://www.microsoft.com/en-us/worklab/work-trend-index/agents-human-agency-and-the-opportunity-for-every-organization
- HBS Working Knowledge reported on October 17, 2025. https://www.library.hbs.edu/working-knowledge/when-ai-joins-the-team-better-ideas-surface