TL;DR: Most builders can’t accurately assess their AI dependency because the tool is integrated into daily work. This 10-question assessment cuts through rationalization by asking what actually happens when you can’t use AI, what you avoid without it, and where friction appears.

The Short Version

You’re productive with AI. Your output is measurable. Your work is good. So how would you even know if you’re dependent? Dependency doesn’t look like laziness—it looks like efficiency. It doesn’t feel like a problem—it feels like leverage.

The difference between using AI as a tool and being dependent on it is a pattern question, not a performance question. This assessment doesn’t measure how much you use AI. It measures what happens when you don’t use it, what you’ve stopped doing since AI arrived, and where you experience friction or anxiety when the tool isn’t available.

Answer honestly. No one is scoring this but you.

The 10-Question Assessment

Question 1: The Friction Test

“When you encounter a problem that would take 30 minutes of thinking but 2 minutes of prompting, how often do you choose the prompt?”

  • A: Almost always (9/10 times)
  • B: Usually, but I think through hard ones first (6/10 times)
  • C: About half and half
  • D: Rarely; I prefer to think through it myself (3/10 times)

What this reveals: The friction test measures whether you’ve lost the habit of struggling with problems. Builders who choose prompting 9/10 times have outsourced cognition. Not because AI is faster—because thinking feels hard now. Your brain has recalibrated what “productive” means.

💡 Key Insight: Dependency starts when thinking becomes the slower option. Once that flip happens, you stop exercising the cognitive muscles that made you a good builder.

Question 2: The Anxiety Gradient

“If you couldn’t access AI tools for one day, how would you feel?”

  • A: Genuinely anxious or stressed
  • B: Annoyed but manageable; I’d get work done slower
  • C: Neutral; I’d adjust and probably be fine
  • D: Maybe relieved for a break

What this reveals: Anxiety is the diagnostic signal. Not frustration at inefficiency—anxiety. If the prospect of losing a tool creates actual stress, your nervous system has learned to depend on it. You’re not just using AI; your threat-detection system is activated by its absence.

Healthy tool users are annoyed by inconvenience. Dependent users experience threat. There’s a neurological difference.

📊 Data Point: Studies on internet addiction show anxiety at access loss is among the earliest detectable markers of dependency, preceding behavioral escalation.

Question 3: The Substitution Gap

“Can you solve a problem with pen and paper that you’d normally solve with AI? Not faster—at all?”

  • A: No; I’d be lost. I don’t even try
  • B: Maybe, but it would take hours and feel clumsy
  • C: I could do it, but it would be half the quality
  • D: Yes, easily. Takes longer but I can still problem-solve

What this reveals: This question detects cognitive atrophy. Dependency doesn’t mean you can’t do the work—it means you’ve lost confidence that you can. The skill hasn’t disappeared; your belief in it has.

Builders who answer “A” are in the danger zone. They’ve internalized that their own thinking is insufficient. This is the psychological anchor of dependency: not incapability, but lost self-trust.

The Progression Questions (4-7)

Question 4: The Morning Protocol

“How does your workday typically begin? (Select one)”

  • A: Open AI tool first, use it to plan or warm up the day
  • B: Email and calendar, then eventually open AI tool
  • C: Dive into the work, reach for AI as needed
  • D: Spend 15-30 minutes thinking/planning before any tools

What this reveals: Your morning routine is your baseline dopamine expectation. If AI is your first action, it’s become a habit, not a tool. You’re not reaching for it to solve a specific problem—you’re reaching for it to start your nervous system.

This is adjacent to checking your phone first thing. It’s not harmful once, but it’s a pattern that reveals where your attention has been anchored.

Question 5: The Avoidance Pattern

“What tasks that you used to do before AI are you now avoiding?”

  • A: Several (design mockups, first-draft writing, code architecture decisions, research synthesis)
  • B: A few (usually the messiest ones—like research)
  • C: Not many; I use AI to speed up tasks I’d do anyway
  • D: I can’t think of any; I’ve just gotten more productive

What this reveals: This is the real diagnostic. Dependency isn’t about using more AI—it’s about doing less of other things. If you’ve stopped doing first drafts, sketching, or research because AI does it, you’ve outsourced skills you might want to keep.

Notice the difference: “AI speeds up tasks I’d do anyway” (tool use) vs. “I avoid tasks because AI exists” (dependency). The former is leverage. The latter is avoidance.

📊 Data Point: Skill loss from disuse can begin in 2-4 weeks for cognitive tasks. If you haven’t done original research, written a first draft, or sketched architecture in months, your cognitive pathways have started rewiring.

Question 6: The Solitude Test

“Can you spend 60 minutes focused on a creative or technical problem without checking an AI tool?”

  • A: No; I’d feel the urge to prompt and interrupt my thinking
  • B: Maybe, but it would feel unnatural
  • C: Yes, but I’d want to use AI afterward to validate
  • D: Yes, easily; it’s actually my preferred way to work

What this reveals: This measures impulsivity and self-interruption. Dependency often looks like reflexive tool-checking, not constant use. You’re in deep thinking, and a part of your brain suggests: “What would AI say about this?” The urge to prompt is intrusive.

This is subtly different from choosing to use AI. Intrusive urges—where the thought of the tool pops into your head unsummoned—are early-stage dependency markers.

Question 7: The Quality Gradient

“How often do you ship work that you generated without reviewing it deeply?”

  • A: Frequently; I trust the AI output
  • B: Sometimes; depends on the task
  • C: Rarely; I always review before shipping
  • D: Never; I’d never ship without my own quality check

What this reveals: Trust in AI-generated work is healthy at a point. But dependency shows up as passive trust—accepting outputs without the rigor you’d apply to your own work. You’re not critically engaging; you’re delegating judgment.

The danger isn’t shipping bad work (you’d notice). It’s shipping mediocre work at scale. You’re optimizing for speed, not quality, and rationalizing it as efficiency.

The Isolation Questions (8-10)

Question 8: The Comparison Anxiety

“When you see someone’s work that’s clearly not AI-generated, what’s your internal response?”

  • A: Slightly anxious; you wonder if your own work is diminished by AI use
  • B: Curious about their process
  • C: Neutral; good work is good work regardless of tools
  • D: Genuinely impressed and interested in how they did it

What this reveals: This measures shame, which is a dependency signal. If you feel anxious about non-AI work being “more real,” you’ve internalized a narrative about your work being less authentically yours. That anxiety drives dependency because you’re trying to prove something through productivity.

Healthy tool users don’t experience shame about tool use. They experience practical tradeoffs. If shame is present, dependency is present.

💡 Key Insight: Shame is the glue that keeps dependent users trapped. It prevents honest conversation, which prevents change.

Question 9: The Offline Scenario

“If you were fully offline (no internet, no AI access) for a week, what would happen to your work?”

  • A: It would slow dramatically or halt; I don’t know how to continue
  • B: It would slow significantly; I’d be less productive
  • C: I’d manage, though with more friction
  • D: I’d be fine; I’d just work differently

What this reveals: This measures resilience and adaptability. Dependent users have organized their entire workflow around AI availability. They don’t have a backup mode. This isn’t necessarily bad until it is—when the tool goes down, they freeze.

Healthy tool users can degrade gracefully. They have a slower mode, a different process, something they fall back to. If you don’t have one, you’re vulnerable.

Question 10: The Honesty Check

“Complete this sentence honestly: ‘I use AI because…’”

  • A: I’m dependent on it; I can’t imagine working without it
  • B: It’s genuinely helpful and I prefer it
  • C: It solves specific problems efficiently
  • D: It’s useful for some tasks; I pick and choose

What this reveals: This is your gut check. Not your rational justification—your honest completion. If you completed it with “I don’t know what I’d do without it,” that’s a dependency signal. If you completed it with a list of specific benefits, that’s tool use.

The difference is binary: Can you articulate why you’re using it (tool), or is it just how you work now (dependency)?

Scoring and What It Means

Count your “A” answers: These are dependency signals.

  • 7-10 As: You’re in dependency territory. Not catastrophic, but the patterns are clear: avoidance, anxiety, lost skills, and reliance. Recovery would require deliberate friction and alternative practices.

  • 4-6 As: Mixed patterns. You have healthy tool use alongside some concerning habits. The good news: you’re still flexible. The work now is noticing which specific patterns are problems.

  • 1-3 As: You’re likely using AI as a tool, not as a dependency. You’re not immune to escalation, but you have the cognitive habits and self-trust to maintain boundaries.

  • 0 As: Either you’re in complete denial, or you have remarkable discipline. Either way, re-read this and be honest.

What This Means For You

The assessment itself isn’t therapeutic. The goal is accurate diagnosis, not reassurance. If you scored in the dependency range, the next step isn’t guilt—it’s decision-making.

Do you want to change? Not because you’re “addicted” (that word carries shame), but because dependency patterns often lead to outcomes you don’t want: lost skills, reduced creativity, outsourced judgment, or eroded confidence.

If the answer is yes, you have options: deliberate friction, time-boxed use, alternative practices, or monitored detoxes. But those only work if you’ve first admitted the pattern exists.

If the answer is no—if you’re fine with the tradeoffs—that’s also legitimate. Some builders genuinely prefer speed to originality. The goal is that it’s a choice you’ve made consciously, not a pattern that chose you.

Key Takeaways

  • Dependency is diagnosed by anxiety at tool loss and avoidance of previous tasks, not by frequency of use
  • Intrusive urges to prompt are early-stage dependency; reflexive tool-checking is behavioral dependency
  • Lost cognitive skills (thinking, first drafts, architecture) are red flags requiring deliberate re-engagement
  • Shame about non-AI work indicates deeper dependency patterns
  • The assessment reveals your relationship with AI; the choice about what to do with that insight is yours

Frequently Asked Questions

Q: Is this assessment actually valid or just click-bait? A: It’s based on addiction and dependency research (internet, behavioral), adapted for AI. It’s not a clinical diagnosis—nothing replaces a therapist. But the patterns are real and recognizable.

Q: What if I scored high? Should I quit AI? A: Not necessarily. Quitting is one option. Breaking the loop through boundaries, friction, and alternative practices is another. The choice depends on your goals and values.

Q: Can you fake this assessment? A: Yes. But why would you? The only person benefiting from honest answers is you.

Not medical advice. Community-driven initiative. Related: Signs You Are Addicted to AI | How to Break Free From AI Addiction | AI Addiction vs. Healthy Use