Jim Atria, Senior Live Online Lead Instructor at MyComputerCareer, spends a good chunk of his time on his weekly “IT Blitz” segment on social media warning students about a trap that’s easy to fall into: letting AI do so much of the thinking that you never build the skill yourself. He put that argument in writing for eCampusNews this month, and it’s a great one for anyone learning or teaching IT.
What Is “Friction Maxxing”?
For years, technology chased one goal: remove friction. One-click checkout, autocomplete, AI that finishes your thought before you do. Now a countertrend has a name: friction maxxing. People are choosing the harder path on purpose (writing by hand, working through a problem without help) because they’ve noticed something. Skip the struggle, and you skip the learning.
Jim’s point isn’t that every convenience is bad. It’s that some challenges need to stay hard, especially in fields like networking, cybersecurity, and systems administration, where troubleshooting under pressure is the actual job.
What the Research Shows
Jim points to a study from Carnegie Mellon, Oxford, MIT, and UCLA that tracked what happens when people use AI on a task. Short-term performance went up. But when researchers took the AI away, those same people performed worse than they had before, and they gave up faster when a problem got hard.
A separate Microsoft and Carnegie Mellon study found something similar in the workplace: the more confidence someone had in an AI-generated answer, the less critical thinking they applied to it. People who trusted their own judgment were more likely to double-check the AI instead of taking its word for it.
How this Impacts IT Training
Take a student learning networking. An AI tool can explain subnetting or generate a configuration example in seconds. Handy, until that student hits a live network outage with no AI in sight and no idea where to start.
Diagnosing a real problem takes judgment, not a correct answer pulled from a chatbot. Jim argues that judgment only comes from doing the work: hitting the wall, working through it, and coming out on the other side with an instinct you didn’t have before. That applies across networking, cybersecurity, systems administration, and software development alike.
How Should Educators Balance AI and Skill-Building?
Jim isn’t arguing for banning AI from the classroom. He’s arguing for sequencing it correctly. His recommendations for instructors:
- Ask students to explain how they reached an answer, not just what the answer is
- Require a genuine troubleshooting attempt before AI enters the picture
- Treat AI as a coach that offers hints, not a machine that hands over solutions
- Grade the process, not only the outcome
- Build labs that force investigation and decision-making under real conditions
Where This Leaves IT Students
AI is going to be part of every IT career from here forward. Students still need to graduate knowing how to use the tech. But the professionals who stand out on the job won’t be the ones who found the fastest shortcut through school. They’ll be the ones who can still diagnose a problem when the shortcut isn’t available.
That’s the thinking behind MyComputerCareer’s IT training programs: hands-on labs, live troubleshooting, and instructors like Jim who make students earn the skill instead of borrowing it from a chatbot.
Curious what that training looks like day to day? Check out our IT training programs and see how we build professionals who can solve a problem, not just prompt one.