We Asked 5 AI Models the Same Question
Layoffs in tech haven't slowed down. 2024, 2025, and now 2026. The cycle continues.
If you're an IT professional actively looking for work, AI can genuinely help. But it can also get you quietly filtered out, ghosted, or flagged if you use it wrong.
So we ran an experiment. We asked 5 frontier AI models the exact same question:
"How should IT professionals find jobs with AI without getting flagged or blacklisted?"
No coaching. No follow-ups. Just one raw question to each model. Here's what they said.
What ChatGPT Said
ChatGPT's core message: "Use AI as a co-pilot, not a clone generator."
The people getting flagged or ignored are the ones submitting obvious, mass-produced applications. The ones getting hired are using AI strategically but still sounding human, specific, and credible.
ChatGPT drew a clear line between AI-generated and AI-assisted:
- ●Bad (AI-ish): "Improved system performance and enhanced scalability."
- ●Good (human + AI assisted): "Reduced API response time by 42% by optimizing Redis caching and rewriting 3 bottleneck endpoints."
The difference? Specifics. Real tools. Real numbers. Real outcomes.
ChatGPT also flagged three resume patterns that recruiters instantly recognize as AI-generated:
- ●Buzzword stuffing ("leveraged synergies," "results-driven professional")
- ●Perfectly structured but vague bullet points
- ●Identical tone across every application
Its advice: use AI to *refine*, not generate from scratch. Inject specifics like tools, metrics, and impact. And add slight imperfections, because natural phrasing beats robotic perfection.
What Claude Said
Claude went straight to what actually gets you ghosted:
- ●Obvious ChatGPT prose like "spearheaded," "leveraged synergies," "dynamic professional," "results-driven leader"
- ●Resume that doesn't match your LinkedIn because ATS systems now cross-reference the two and inconsistencies trigger review
- ●Cover letters with no specificity because screeners can smell a template from across the room
- ●Mass-applying with zero customization because velocity flags exist in some systems
- ●Keyword stuffing because cramming job description terms without context reads as manipulation to both ATS and humans
Claude's bottom line: the real risk isn't AI detection. It's sounding generic when you've been told to "tailor your resume." Every AI-polished resume starts to sound the same, and hiring managers notice.
What Claude Code Said
Claude Code framed AI as a force multiplier for quality, not quantity.
What gets you flagged:
- ●Submitting high volumes of near-identical applications (easy-apply spam)
- ●Copy-pasting AI output verbatim into assessments or screening questions
- ●Using bots to auto-apply (platforms like LinkedIn actively detect and penalize this)
- ●Keyword-stuffing resumes with invisible text (white-on-white) to game ATS filters
What actually works:
- ●Use AI to identify skill gaps and tailor your resume to specific job descriptions
- ●Leverage AI to research companies and prepare for interviews
- ●Use AI to practice technical interviews (mock coding, system design scenarios)
- ●Let AI help you write networking messages, then personalize them
- ●Build portfolio projects that demonstrate you can work with AI tools effectively, because that is a selling point in IT right now
Claude Code's key principle: platforms flag volume and pattern. They reward relevance and engagement. A few well-targeted, personalized applications outperform hundreds of spray-and-pray submissions.
What Grok Said
Get the Weekly IT + AI Roundup
What changed this week in NinjaOne, ServiceNow, CrowdStrike, and AI. One email, every Monday.
No spam, unsubscribe anytime. Privacy Policy
Grok pulled real-time data from X/Twitter and laid out a 6-step framework:
1. Adopt the right mindset. Treat AI like spellcheck or a junior assistant. Always start with your own draft (experience, metrics, projects) and use AI to enhance. Never fabricate skills, projects, or experience, because tech interviews (coding, system design, behavioral) catch this instantly.
2. Job discovery and company research. Use AI to analyze job descriptions: "Identify the top 5 technical skills and scalability challenges in [your tech stack]." This positions you as a strategic consultant rather than a generic applicant.
3. Build ATS-friendly, humanized resumes. Paste the job description into AI, extract keywords, rewrite for impact and quantity. Then heavily edit in your voice. Add unique metrics from your projects, GitHub links, or anecdotes. Vary sentence structure, read aloud, and inject personality. Generic, polished output gets binned fast because 80% of hiring managers dislike obvious AI resumes.
4. Handle cover letters carefully. Many recruiters explicitly say don't use AI to draft from scratch. Better approach: write your own short, personalized version first, then feed in your research. Add a real anecdote. Keep it concise. Skip entirely if you can't make it authentic.
5. Networking and referrals. This is the best way to bypass all screens. AI-optimized LinkedIn profiles are fine. Prioritize referrals, GitHub contributions, and community engagement. These often skip AI/ATS screening entirely and have 4x higher success rates.
6. Interview prep and follow-through. Use AI for mock interviews. Practice responses aloud in your own words. Never read AI scripts live. Post-interview, use AI to draft thoughtful thank-yous tied to conversation specifics.
Grok's bottom line: 80% human effort + 20% AI polish. Target 5-10 tailored applications per week, not 100 spray-and-pray.
What Gemini Said
Gemini cited real data: 74% of hiring managers now claim they can spot AI-written cover letters. If you use one, you risk being flagged for "low effort."
Its framework: use AI to audit, not author.
| Use AI For | Avoid |
|---|---|
| Quantifying impact: "Rewrite this bullet point to include specific metrics like latency reduction or cost savings." | Drafting from scratch: Don't let AI write your summary. It will sound like 1,000 other candidates. |
| ATS gap analysis: Paste your resume and the JD. Ask "What technical skills am I missing that this JD emphasizes?" | Keyword stuffing: Adding hidden keywords in white text or list-dumping tools you haven't actually used. |
| Tone checking: "Does this sound too formal or too casual for a startup engineering culture?" | Template overuse: Using the standard ChatGPT table or bullet structure that recruiters see 50 times a day. |
Gemini also warned about the "auto-apply" trap. Mass-blasting 500 applications via a bot often results in an automatic shadow-ban by high-intent filters. Quality over quantity: use AI to find the 5-10 most relevant jobs and customize 100% of the content.
Its advice for standing out: focus on "the weird." AI is great at the average. To differentiate, emphasize your unique edge cases, the strange bugs you solved, the non-linear career paths you took.
Where All 5 Agreed
Despite getting no context about each other's answers, every model converged on the same core principles:
- ●Never let AI write your resume from scratch. Use it to improve what you've already written.
- ●Specifics beat polish every time. Real metrics, real tools, real outcomes.
- ●Mass-applying with bots gets you shadow-banned. Platforms detect velocity and pattern.
- ●Referrals bypass everything. Networking is still the highest-ROI job search activity.
- ●Your real stories are your competitive advantage. Your 2AM outage fixes, your tradeoff decisions, your failures and recoveries. AI can't invent those.
Where They Disagreed
| Model | Unique Angle |
|---|---|
| ChatGPT | Natural imperfections are better than robotic perfection. Add slight roughness to your writing. |
| Claude | Resume and LinkedIn consistency matters. ATS systems cross-reference them now. |
| Claude Code | Build portfolio projects showing you work WITH AI tools. That's a selling point. |
| Grok | Referrals have 4x higher success rates. Networking beats optimization. |
| Gemini | 74% of hiring managers spot AI cover letters. Write them yourself or skip them entirely. |
The Bottom Line
AI is not the cheat code. AI is the prep work. Your experience is the product.
Every model said the same thing in different words: the people who get flagged are the ones who let AI do all the thinking. The people who get hired are the ones who use AI to sharpen what they already know.
If you're mid-search right now, try all 5 models. They're free to start. Each one catches something the others miss.
*Before you start applying, run your resume through our free Resume Compatibility Checker to see how well you match the job description.*
*Want a step-by-step plan? Our free AI Job Search Playbook covers everything from beating ATS filters to getting found by hiring managers. 6 modules, completely free.*
*Not sure which AI fits your workflow? Take our free quiz and find out in 2 minutes.*