# The Principal's Script for the "We're Not Forcing This, But..." Conversation **By Dan Cumberland** · Published August 17, 2026 · Categories: AI Strategy ![Featured image for The Principal's Script for the "We're Not Forcing This, But..." Conversation](https://cms.dancumberlandlabs.com/uploads/featured_image_45c0eec2fa.webp) Illustration: Dan Cumberland Labs with Gemini. > Saying "we're not forcing this, but..." signals exactly what it intends to hide: adoption is expected. Most AEC firm principals understand the business case... Saying "we're not forcing this, but\.\.\." signals exactly what it intends to hide: adoption is expected\. Most AEC firm principals understand the business case for AI\. What many don't have is language for a conversation that's honest about the expectation while preserving the autonomy that makes adoption genuine— not grudging\. The problem isn't whether AI is worth pursuing\. It's how the conversation gets framed\. When people feel that choice has been taken away, research shows they don't just hesitate— they actively resist\. And that resistance is what turns a straightforward AI adoption conversation into a morale problem\. This article provides the behavioral framework and the actual language to bridge that gap\. According to Bluebeam's 2025 AEC industry survey[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1), 68% of early adopters have saved at least $50,000 using AI\. The business case is solid\. But Training Industry research[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2) found that up to 70% of change initiatives fail due to employee pushback— meaning the business case alone isn't enough\. ## What the Data Says About AEC's AI Moment AEC firms that adopted AI early are saving real money\. Those that haven't are watching their peers widen the gap\. That's the business reality principals are navigating when they call this meeting\. Only 27% of AEC firms currently use AI for automation, problem\-solving, or decision\-making— but those that do are pulling ahead fast[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1)\. Early adopters report 68% saving at least $50,000, and 46% saving between 500 and 1,000 hours annually[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1)\. Those aren't marginal gains\. That's structural advantage that compounds over time\. The market momentum is clear\. A full 94% of AEC companies currently using AI plan to increase their investment in the coming year[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1)\. And alongside the ROI case, principals are navigating legitimate barriers: 42% of firms cite data security concerns, 33% flag cost and complexity, and 69% worried about regulatory uncertainty at the time of the survey[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1)\. The principal's job isn't to pretend these barriers don't exist— it's to acknowledge them honestly while making the case that the cost of waiting is now higher than the cost of moving\. ```html-table
Early AEC Adopters ReportCurrent AEC Majority
68% saved $50,000 or more73% not yet using AI for problem-solving
46% saved 500–1,000 hoursWatching competitors compound the gap
94% increasing investment next yearStill evaluating the business case
``` ## Why Staff Resistance Has Nothing to Do With AI Staff resistance to AI usually isn't about AI\. It's about job security, skill obsolescence, and feeling like choices are being made for them\. When principals understand what's actually driving resistance, the conversation changes\. Research is clear on what staff actually fear: their skills becoming obsolete and their career trajectories shifting in ways they can't predict[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2)\. That's not irrational— it's a reasonable response to genuine uncertainty\. And when that uncertainty meets a top\-down announcement, the situation gets worse, not better\. Peer\-reviewed research in the European Journal of Information Systems[3](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-3) found that individuals denied choice in technology adoption are more likely to engage in opposition behavior— even when they can see the tool's value\. That's not stubbornness\. That's how autonomy works\. The "soft mandate" creates exactly the problem it's trying to avoid: signaling expectation while framing adoption as optional triggers the resistance it was designed to prevent\. What staff are actually worried about: - Whether their expertise will still matter once AI handles the "easy" parts of their work - Whether they'll have enough time and real support to learn something new - Whether making mistakes during the learning curve will be held against them These aren't irrational fears, and addressing them isn't a soft detour from the business case— it's the path through it\. And this isn't a small number of people\. Research cited through McKinsey's analysis[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4) found that 31% of U\.S\. knowledge workers actively work against company AI initiatives\. According to Training Industry[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2), 70–80% of AI projects fail to deliver expected benefits— not because the technology doesn't work, but because users don't genuinely adopt it\. The limiting factor isn't the tool\. It's the human relationship with the tool\. ## The Framework: Inspire First, Require Second The most effective AI adoption strategy isn't a mandate\. It's a sequence: build desire before you set expectations\. McKinsey research and behavioral science research converge on the same insight— inspire first, require second\. Forbes Technology Council contributor Paul Deraval[5](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-5) put it plainly: "Inspire first\. Require second\." The order isn't optional— it's the mechanism\. Top\-down rollouts fail with AI because the human stakes— career anxiety, skill erosion, trust— demand a different approach[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2)\. The ADKAR Model, developed by change management research firm Prosci[6](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-6), maps the psychological journey from first hearing about a change to genuinely believing in it\. Applied to AI adoption in an AEC firm, each stage requires something specific from the principal: ```html-table
ADKAR StageWhat It Means for AI AdoptionWhat the Principal Does
AwarenessWhy is AI adoption a priority for this firm right now?Share the business context honestly— data, competitive pressure, client expectations
DesireHow does AI make this person's work better?Connect to individual career relevance, not just firm efficiency
KnowledgeWhat specific tools will we use, and how will I learn them?Name the tools; commit to a training plan before asking for adoption
AbilityWhat support is available when I get stuck?Office hours, a Slack channel, identified early users to learn from
ReinforcementHow will early wins be recognized?Celebrate examples publicly; connect adoption to career outcomes
``` The table above gives you the diagnostic— how to think about where a person is in the adoption journey and what they need from you at each stage\. But the *design*— how you structure the workflow itself to support adoption— is where nudge theory fills a gap the ADKAR model doesn't address\. WalkMe's research[7](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-7) describes nudge theory as "using small, gentle pushes to guide people toward better decisions while preserving individual freedom\." In practice, this means designing the workflow so AI is the easiest path— not banning alternatives, just making the AI\-enabled approach the default\. The FEAST framework[7](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-7) makes this concrete: Fun \(gamify early wins\), Easy \(reduce friction\), Attractive \(show the career upside\), Social \(let peers model it\), Timely \(introduce AI when a real pain point is visible\)\. But one guardrail matters: nudge approaches only work when they're transparent\. Using behavioral design without disclosure creates the exact trust problem you're trying to solve[7](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-7)\. That's worth stating clearly in the conversation itself\. For principals working on [building an AI\-ready culture](/blog/building-ai-culture), the framework isn't the hard part\. The hard part is having the conversation that starts it\. ## The Conversation: What AEC Firm Principals Should Actually Say The core message is this: AI amplifies your expertise— it doesn't replace it\. That's not spin\. It's accurate\. The question is how to deliver it in a way your team can believe\. Start with business context specific to your firm— not a generic AI evangelism moment\. Here's how a principal might open this conversation: > "Here's where we are: firms that adopted AI early are saving real time and money, and we're at a point where waiting has a cost\. I'm not here to tell you this will be easy or that I have every answer\. What I do know is that I want us to figure this out together\. My question for you today isn't 'will you use AI?'— it's 'what would make this actually useful for your work?'" Forbes Technology Council research[5](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-5) frames the core message clearly: position AI as amplifying expertise, not replacing workers\. "The routine parts of your work will move to AI\. That frees you for the judgment, client relationships, and complex problem\-solving that define your value to this firm\." This isn't a talking point— it's what happens when [AI implementation](/services/ai-implementation) is designed well\. Acknowledging uncertainty honestly builds more trust than false certainty\. Training Industry research[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2) calls this "clarity is kindness"— even when the details remain unclear\. "I don't have all the answers about how this will change your specific role\. What I know is that the firms ignoring this are falling behind\." That's an honest thing to say\. Staff can work with honesty in ways they can't work with performance\. And the conversation shouldn't end when the principal finishes talking\. Two\-way communication is what converts a compliance directive into genuine adoption[2](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-2)\. Build in: - Dedicated Q&A time during the initial conversation - A follow\-up channel \(Slack, recurring office hours\) open for the next 90 days - An explicit invitation: "Tell me what you're worried about— that information helps me design this rollout better" PwC's "My AI" initiative[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4) is instructive here\. PwC structured AI adoption around peer activators— internal champions who combined hands\-on experimentation with structured tools\. The result: adoption became social and self\-reinforcing rather than mandated from above\. Avoid any framing that implies AI will be used to monitor performance\. "We'll be able to see who's using it" triggers exactly the surveillance anxiety that drives underground resistance\. The conversation should feel like an invitation, not the start of an audit\. If tracking is genuinely needed for operational reasons, name it upfront: "We'll track which tools are being used to figure out where training is needed— not to evaluate individual performance\." If your firm needs formal policy alongside the cultural conversation, an [AI governance framework](/blog/ai-governance-strategy) can structure that layer separately\. ## Handling the Hard Questions Staff won't ask "What's your AI governance strategy?" They'll ask "Is this going to take my job?" Have a real answer ready\. **Objection: "Will this replace my job?"** Answer: "No— but the work will shift\. AI handles the formulaic, repetitive parts\. You focus on judgment, client relationships, and the complex problems that require expertise\. That's higher\-value work, not less of it\." The legal services parallel is instructive\. AI adoption among legal professionals more than doubled in a single year, reaching 69% using generative AI tools for work in 2025[8](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-8)\. The attorneys at those firms didn't see role elimination— they saw reallocation toward the work that requires a human\. Professional services firms that pair AI rollout with honest role redefinition consistently see genuine adoption, not compliance theater\. Some professionals who started out requesting refunds on AI tools they found unusable have gone on to build their own AI solutions and become their firm's most effective AI advocates— because the conversion came from genuine understanding and autonomy, not top\-down pressure\. That arc is repeatable when the conditions are right\. Gartner's December 2025 HR survey[9](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-9) found that 65% of employees are already excited about using AI at work\. You're not fighting a hostile room\. You're managing anxiety, not opposition\. **Objection: "When will I have time to learn this?"** Answer: "We're building that time in\. This isn't learn\-it\-on\-your\-own\. Here's what training looks like\." If the training plan doesn't exist yet, this objection isn't the employee's problem— it's exposing a gap in the rollout\. Don't promise "support" in the abstract\. Name the specific resources before you ask people to adopt\. **Objection: "What if I make a mistake?"** Answer: "We're treating the first 90 days as a learning phase to figure out what works for our firm's workflow— not as a referendum on whether we adopt\. Your job is to experiment, share what you find, and tell us what doesn't work\." This reframes adoption as a shared experiment rather than a compliance test\. Resistance often stems from adversarial framing and opaque systems, not from opposition to innovation itself[10](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-10)\. A learning\-phase framing removes the punitive shadow from the process\. ## What Comes After the Conversation One meeting doesn't drive AI adoption\. The 90 days after the meeting— what structures you build, who you identify as internal champions, and how you measure success— determine whether people genuinely adopt or just comply\. McKinsey research[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4) highlights the superuser strategy: identify 1\-2 staff members who show early enthusiasm or curiosity\. Give them explicit support, time to experiment, and a platform to share what they learn\. They become change agents the team trusts more than the principal— because the trust is peer\-to\-peer\. The distinction between compliance and adoption is real and measurable\. According to McKinsey[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4), only about 1 in 10 employees strongly believe AI changes how their work gets done, while 45% report using AI regularly\. That gap is compliance masquerading as adoption\. McKinsey[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4) also found that 64% of organizations lack robust KPIs for tracking AI initiatives— which means most firms can't tell when they've achieved adoption versus surface usage\. But measuring compliance is easier than measuring adoption\. ```html-table
Compliance SignalsAdoption Signals
Uses AI when observed or when askedReaches for AI before starting a project deliverable or spec document
Reports tool usage in check-insUses AI to draft submittals, meeting summaries, or RFI responses before asking for help
Completed assigned trainingAsks "can I use AI for this?" before starting new project phases or proposal work
Hasn't raised complaints about the toolShares workflow shortcuts with project team members without being prompted
``` Follow\-up cadence for the first 90 days: - Monthly check\-ins— not performance audits, but "What's working? What's not? What should we do differently?" - Public recognition when early wins surface - A feedback loop that actually modifies the rollout based on what you hear For more on how to track this over time, [measuring AI adoption success](/blog/measuring-ai-success) requires behavioral signals— not just login data\. ## Frequently Asked Questions ### What should AEC firm principals say when introducing AI to their team? Lead with the specific business context for your firm— early AEC adopters are saving $50,000\+ and 500–1,000 hours annually[1](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-1)— position AI as capability amplification rather than workforce reduction, and build a two\-way dialogue rather than a one\-way announcement\. The core message: "AI amplifies your expertise\. We're adopting this to make you more capable, not to make you replaceable\." ### Why do soft mandates for AI adoption fail? Soft mandates signal an expectation while framing adoption as optional— which creates exactly the psychological conflict they're designed to avoid\. Peer\-reviewed research[3](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-3) shows that when employees feel they've been denied a choice, they're more likely to engage in opposition behavior, even against tools they might otherwise value\. The solution isn't vagueness about expectations— it's building genuine desire before setting them\. ### How should principals respond to "Will AI replace my job?" The honest answer is: "No— but the work will shift\. AI handles the formulaic work\. You focus on judgment and client relationships, which are more valuable\." Professional services firms in legal and consulting that deployed AI early saw reallocation toward higher\-value work, not role elimination[8](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-8)\. The firms whose staff believed this were the ones who demonstrated it over time, not just asserted it once\. ### What is the ADKAR model for AI adoption? ADKAR is a five\-stage change management framework from Prosci[6](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-6): Awareness \(why this matters\), Desire \(what's in it for me\), Knowledge \(how to use it\), Ability \(practicing it\), and Reinforcement \(sustaining it\)\. Applied to AI adoption, each stage requires different actions from the principal\. The table in the framework section above maps those actions specifically for AEC context\. ### How do you measure real AI adoption versus compliance? Compliance looks like usage when observed\. Adoption looks like reaching for AI when no one's watching and when it makes the work genuinely easier\. McKinsey research[4](/blog/blog-the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation#ref-4) found that only 1 in 10 employees who use AI regularly actually believe it changes how their work gets done\. Measure behavioral integration— are people changing how they approach client work?— not just tool logins\. ## The Conversation Is the Leverage Point The principal's job isn't to convince people that AI is good\. It's to create the conditions where they discover that for themselves\. No matter the question, people are the answer\. The mandate, the tool, the rollout plan— none of it works if the people on your team don't believe in it\. The conversation you have \(and how you have it\) is where that belief starts or doesn't\. Your own relationship with AI sets the tone more than any script\. Staff watch whether you use it, whether you acknowledge what you don't know, and whether you're learning alongside them\. The principal who says "I'm still figuring parts of this out too" will get further than the one who performs certainty they don't have\. If your firm is navigating AI adoption for the first time and you want an outside perspective on how to structure the conversation and the rollout, [Dan Cumberland Labs works with AEC and professional services firms](/services/ai-implementation) to design adoption programs built around genuine belief— not compliance theater\. ## References 1. Bluebeam, "New Bluebeam Report Shows Early AI Adopters in AEC Seeing Significant ROI Despite Uneven Adoption" \(2025\)— [https://press\.bluebeam\.com/2025/10/new\-bluebeam\-report\-shows\-early\-ai\-adopters\-in\-aec\-seeing\-significant\-roi\-despite\-uneven\-adoption/](https://press.bluebeam.com/2025/10/new-bluebeam-report-shows-early-ai-adopters-in-aec-seeing-significant-roi-despite-uneven-adoption/) 2. Training Industry, "How to Overcome 4 Common AI Adoption Resistance Scenarios" \(2026\)— [https://trainingindustry\.com/articles/artificial\-intelligence/how\-to\-overcome\-4\-common\-ai\-adoption\-resistance\-scenarios/](https://trainingindustry.com/articles/artificial-intelligence/how-to-overcome-4-common-ai-adoption-resistance-scenarios/) 3. European Journal of Information Systems, "Do I Really Have To? User Acceptance of Mandated Technology" \(2004\)— [https://www\.tandfonline\.com/doi/abs/10\.1057/palgrave\.ejis\.3000438](https://www.tandfonline.com/doi/abs/10.1057/palgrave.ejis.3000438) 4. McKinsey, "McKinsey State of AI 2025 / Reconfiguring Work in the Age of Gen AI" \(2025\)— [https://www\.gend\.co/blog/mckinsey\-state\-of\-ai\-2025\-key\-findings\-what\-to\-do](https://www.gend.co/blog/mckinsey-state-of-ai-2025-key-findings-what-to-do) 5. Forbes Council, "Inspire Versus Require: The New Mandate For AI Leadership" \(2026\)— [https://www\.forbes\.com/councils/forbestechcouncil/2026/02/05/inspire\-versus\-require\-the\-new\-mandate\-for\-ai\-leadership/](https://www.forbes.com/councils/forbestechcouncil/2026/02/05/inspire-versus-require-the-new-mandate-for-ai-leadership/) 6. Prosci, "ADKAR Change Management Model" \(2025\)— [https://www\.prosci\.com/blog/adkar\-model](https://www.prosci.com/blog/adkar-model) 7. WalkMe, "What is Nudge Theory? Examples & Best Practices" \(2025\)— [https://www\.walkme\.com/blog/nudge\-theory/](https://www.walkme.com/blog/nudge-theory/) 8. AllRize, "Legal Technology and AI Adoption Report 2025" \(2025\)— [https://allrize\.ai/wp\-content/uploads/2025/10/AllRize\_Legal\_Industry\_Technology\_and\_AI\_Adoption\_Report\_2025\.pdf](https://allrize.ai/wp-content/uploads/2025/10/AllRize_Legal_Industry_Technology_and_AI_Adoption_Report_2025.pdf) 9. Gartner, "Gartner HR Survey Finds 65% of Employees Are Excited to Use AI at Work" \(2025\)— [https://www\.gartner\.com/en/newsroom/press\-releases/2025\-12\-16\-gartner\-hr\-survey\-finds\-65\-percent\-of\-employees\-are\-excited\-to\-use\-ai\-at\-work](https://www.gartner.com/en/newsroom/press-releases/2025-12-16-gartner-hr-survey-finds-65-percent-of-employees-are-excited-to-use-ai-at-work) 10. Kognitos, "The Real Cause of AI Resistance and How to Solve It" \(2025\)— [https://www\.kognitos\.com/blog/ai\-resistance/](https://www.kognitos.com/blog/ai-resistance/) --- **How we made this article:** We use AI in our research and writing so our small team can share more of what we learn. We verify the sources and take responsibility for every article we publish. [Read how we use AI.](https://dancumberlandlabs.com/how-we-use-ai/) --- ## About the author **Dan Cumberland** — Founder, Dan Cumberland Labs Dan Cumberland helps engineering and construction firms see where they stand with AI and decide what to build first. He created Pacemark, the AI maturity model behind that work, from research on more than 300 companies. - Take the assessment: https://pacemark.ai/signal/assessment/?track=aec&utm_source=dcl-site&utm_medium=link&utm_campaign=pacemark-assessment - Book a call: https://book.dancumberland.com/ai-strategy --- Source: https://dancumberlandlabs.com/blog/the-principal-s-script-for-the-we-re-not-forcing-this-but-conversation/