Why Your AI Strategy Is Backwards
The Lawyer Who Got Caught Twice
In December 2025, Larry Mason made a mistake that would cost his firm $59,500 and his reputation considerably more. The Chicago attorney submitted a brief to federal court containing twelve legal citations. Every single one was fake.
ChatGPT had invented case names, fabricated quotes, and even created internal citations that referenced other fictional cases. The Chicago Housing Authority’s attorneys couldn’t find a single citation in any legal database. When they flagged this to the court, the judge gave Mason a chance to verify his sources.
Here’s where it gets worse: Mason went back to ChatGPT and asked if the cases existed. The AI confidently assured him they “indeed exist” and “can be found in reputable legal databases such as LexisNexis and Westlaw.”
So Mason submitted them again.
The court wasn’t amused. Mason and his firm, Goldberg Segalla, faced $59,500 in sanctions. His name became a cautionary tale in legal AI circles—the attorney who trusted AI not once, but twice, despite clear warning signs.
Mason isn’t an outlier. He’s a preview.
Every industry has a version of this story waiting to happen. The consultant who presents AI-generated analysis without verification. The customer service director who deploys a chatbot that makes promises the company can’t keep. The CEO who invests six figures in AI tools because competitors are posting about it on LinkedIn.
The pattern is always the same: professionals who would never trust a brand-new employee’s unsupervised work somehow trust AI output without question. The same people who demand multiple reviews before approving a press release will paste AI-generated text directly into client deliverables.
This isn’t stupidity. It’s psychology. And understanding why smart people make predictable AI mistakes is the first step to not becoming one of them.
The 95% Failure Rate
Let me give you a number that should make you uncomfortable: 95% of enterprise AI pilots deliver zero measurable return.
Not “modest return.” Not “below expectations.” Zero.
This comes from MIT’s NANDA study, which tracked generative AI investments across major enterprises. The vast majority are getting nothing back from their AI investments.
How is this possible? How can organizations deploy AI everywhere and see nothing for it?
The answer is simple: they’re doing it backwards.
Most organizations approach AI by asking these questions, in this order:
- Which AI tool should we buy?
- What can we automate?
- How fast can we deploy?
These seem like reasonable questions. They’re also completely wrong.
The right questions are:
- Where do humans need better information to make decisions?
- How will we verify AI output before acting on it?
- How does the AI earn expanded responsibility over time?
The first approach treats AI as a solution looking for a problem. The second treats it as a new team member who needs to prove themselves.
About 67% of CEOs admit that FOMO—fear of missing out—drives their AI investment decisions. They see competitors announcing AI initiatives. They read headlines about productivity gains. They feel pressure to act before they’ve identified what problem they’re solving.
That pressure leads to the 95% failure rate. Organizations deploy AI before understanding what success would look like.
Why Smart People Make Predictable Mistakes
Here’s the uncomfortable truth: even when organizations ask the right questions, human psychology works against them.
Five cognitive traps explain why AI adoption fails even among intelligent, well-intentioned professionals. These aren’t character flaws. They’re how human brains work. Understanding them is the first step to counteracting them.
Trap 1: Automation Bias
AI output looks authoritative. It’s formatted professionally. It uses confident language. There’s no hedging, no “I’m not sure about this,” no visible uncertainty.
The same analysis from a human colleague would trigger healthy skepticism. From AI, it triggers acceptance.
Larry Mason fell into this trap. The fake legal citations looked like real legal research—proper formatting, confident citations, internal consistency. His brain processed the professional appearance as a signal of accuracy. It wasn’t.
Trap 2: The Reversed Dunning-Kruger Effect
Here’s a finding that should concern anyone who considers themselves “good at AI”: research from Aalto University found that higher AI literacy leads to more overconfidence, not less.
The more comfortable you are with AI, the more you trust its output. Your prompting skills make you feel protected from errors. They don’t.
This means your most tech-savvy employees may be your biggest AI liability. They’re the ones most likely to skip verification because they believe they know how to get accurate results.
Trap 3: The Urgency Trap (FOMO)
When everyone around you is deploying AI, waiting feels like falling behind. CEOs return from conferences buzzing about AI announcements. Competitors post LinkedIn content about their “AI-powered” operations. Customers ask if you’re “investing in AI.”
This pressure creates a “deploy now, fix later” mentality. Organizations launch AI pilots without clear success criteria. When those pilots inevitably struggle, no one knows whether they failed or were never properly defined.
Trap 4: Sunk Cost and Pilot Purgatory
Here’s a pattern I see constantly: A pilot shows promising early results. Leadership expands scope. Problems emerge. Rather than stopping, the organization “iterates” indefinitely.
The pilot never scales. It never dies. It just consumes resources while everyone waits for it to finally work.
This is how $180,000 pilots become $2.3 million money pits. Each additional investment feels smaller than abandoning the cumulative spend. Organizations can’t kill initiatives that aren’t working.
Scale Doesn’t Protect You
You might think these traps only catch small operators or unsophisticated users. The evidence says otherwise.
Air Canada deployed a chatbot that told a grieving customer he could apply for bereavement rates within 90 days of ticket purchase. The customer booked a flight, flew to a funeral, then discovered the chatbot was wrong—Air Canada’s actual policy doesn’t apply bereavement rates to completed travel.
When the customer sued, Air Canada tried an innovative defense: the chatbot was “a separate legal entity that is responsible for its own actions.” The tribunal was unimpressed. Air Canada lost, and the case established a principle that matters for every organization: companies remain liable for their AI tools’ mistakes. “The AI said it” is not a legal defense.
Amazon spent three years developing an AI recruiting tool before discovering it had learned to discriminate against women. The system penalized resumes containing the word “women’s” (as in “women’s chess club”) and favored verbs common on male engineers’ resumes. Amazon’s engineers tried to fix the bias but couldn’t ensure the system wouldn’t learn other discriminatory patterns. They scrapped the entire three-year project.
A Stanford study found that 90% of AI-related court errors come from solo practitioners or small firms. The pattern is the same whether you’re a Fortune 500 or a one-person shop: AI deployed without verification produces predictable failures.
The difference isn’t sophistication. It’s whether you treat AI output as finished work or work that needs review.
The Intern Model: A Framework That Actually Works
Here’s where we get practical.
The most effective mental model for AI isn’t “powerful tool” or “automation engine” or “the future of work.” It’s much simpler: treat AI like an eager, capable intern.
Think about how you’d work with a talented new hire:
- They work fast — interns are often enthusiastic and quick
- They produce confident output — they don’t always know what they don’t know
- They’re sometimes completely wrong — they lack experience to catch their own errors
- They require supervision — no one expects to use intern work without review
- They earn trust incrementally — they start with small tasks and graduate to bigger ones
This perfectly describes AI’s current state.
The intern model has four principles. Each one directly counteracts the cognitive traps that cause AI failures:
1. Clear Tasks. Never say “handle this.” Define scope, criteria, and boundaries. Tell the AI exactly what you need, exactly what format you want it in, and exactly what constraints apply. This forces you to think through what you’re actually trying to accomplish before you deploy AI—counteracting the FOMO-driven rush to “just try it.”
2. Review Before Shipping. Nothing goes to production, customers, or courts without human eyes. Every output gets reviewed with appropriate scrutiny for its stakes. This creates a mandatory pause that interrupts automation bias—the tendency to accept professional-looking output without verification.
3. Incremental Trust. Start small. Expand responsibility as the AI demonstrates reliability in your specific context. Don’t assume performance in one area transfers to another. This prevents the overconfidence that comes from AI literacy—even experts need to verify AI works for their specific use case.
4. Feedback Loops. Correct mistakes explicitly. Track patterns. Adjust your approach based on what works and what doesn’t. This creates natural exit points that prevent pilot purgatory—if the feedback consistently shows problems, you have clear evidence to stop rather than escalate.
The framework maps directly to the traps:
| Cognitive Trap | Intern Model Countermeasure |
|---|---|
| Automation bias | “Review before shipping” creates a mandatory pause |
| Reversed Dunning-Kruger | Incremental trust prevents overconfidence |
| Urgency trap (FOMO) | Clear tasks force scoping before deployment |
| Sunk cost / pilot purgatory | Feedback loops create natural exit points |
| Authority heuristic | Treating AI as “junior” reduces default trust |
The pattern works at every scale—from solo practitioners to Fortune 500 companies. The question is whether you’ll adopt it proactively or learn it the hard way.
How This Plays Out in Practice
The cognitive traps hit everyone, but they manifest differently depending on your position. Here’s what the backwards approach looks like—and what the fix looks like—in two common situations.
The Customer Service Director
Rachel runs a 28-person customer service team at a regional insurance company. Her CEO returned from a conference excited about AI chatbots, and within weeks, Rachel had budget approval for a chatbot promising 40% call deflection.
She fast-tracked implementation—four weeks from contract to go-live during open enrollment. The chatbot looked impressive in demos. It passed basic testing.
Then it told three policyholders their cancer treatments were covered under plans that explicitly excluded them. One scheduled surgery based on this information.
The chatbot had been trained on marketing materials (which emphasized coverage) rather than policy documents (which detailed exclusions). Rachel’s team trusted the professional-looking output without verifying against actual policy language.
The intern model fix: Rachel should have defined clear task boundaries: “Answer FAQs about enrollment dates only. Coverage questions go to human agents—period.” She should have required senior reps to review 100 chatbot responses against policy documents before any customer interaction. Trust in coverage discussions could come later, after proving reliability on simpler tasks.
The Small Company CEO
David runs a 34-person precision machining company. At a trade show, every vendor booth featured AI prominently. His competitors were posting about “AI-powered quality control.” His largest customer asked if he was “investing in Industry 4.0.”
Within 90 days, David committed to three AI initiatives totaling $127,000: predictive maintenance, computer vision quality inspection, and AI-enhanced quoting.
Results: The predictive maintenance system needed 18 months of historical data—David had 6 months. 89% false-positive rate. Canceled at a $55,000 loss. Computer vision worked for 3 of 7 part families, required a full-time exception handler, ran at negative ROI. AI quoting actually worked well—$31,000 in time savings.
Net result: David’s 1.5% of revenue AI investment became a 0.8% of revenue loss.
The intern model fix: David should have started with quoting alone—clear ROI, simple verification, $18,000 investment. Prove value over six months. Use that success to fund a pilot of one additional system. Collect 18 months of machine data before considering predictive maintenance. Sequential, verified implementation beats parallel deployment every time.
Common Objections
“This seems like a lot of overhead. Won’t all this review slow us down?”
Review takes time. Fixing errors after they reach customers, courts, or public view takes much more time. Rachel spent 40+ hours reviewing chatbot interactions after a failed deployment. If she’d spent 10 hours on pre-deployment verification, she’d have saved 30 hours and a lawsuit.
“Our team is experienced with AI. We don’t need training wheels.”
Remember the reversed Dunning-Kruger effect: higher AI literacy correlates with more overconfidence, not less. Your most experienced AI users may need these frameworks most.
“We’re just using AI for low-stakes tasks.”
Every task that touches customers, data, or business decisions has stakes. The question is whether you’ve correctly identified them. The Air Canada chatbot seemed low-stakes until a customer relied on its bereavement policy advice.
“We can’t slow down—our competitors are moving fast.”
Your competitors are also failing at a 95% rate. Moving fast in the wrong direction isn’t an advantage. The organizations that will win are the ones that build reliable AI processes, not the ones that deploy most quickly.
Your Monday Morning Action Item
Before this week ends, run this test: look at the last three times you or your team used AI output in actual work—an email, a report, a customer communication, anything.
For each one, ask:
- Did anyone review this output before it went live?
- Would we have trusted this output from a new employee without review?
- What would have happened if this output contained a significant error?
If your honest answers are “no,” “yes,” and “something bad,” you’ve identified where to start.
The intern model begins with awareness: recognizing the gap between how we treat human work and how we treat AI work. Once you see it, you can’t unsee it.
Think back to Larry Mason, the attorney who submitted fake citations twice. What if he’d treated ChatGPT like an intern?
“Hey, I need to verify these citations before filing. Can you confirm each one exists in Westlaw?”
ChatGPT would have still confidently assured him. But with the intern frame, Mason might have thought: “Wait—I wouldn’t trust an intern’s assurance that their research is correct. I’d check it myself.”
That pause—that moment of appropriate skepticism—is what the intern model creates. It’s the difference between a $59,500 sanction and a brief that withstands scrutiny.
And once you build that pause into everything you do with AI, you’re ready for Chapter 2.