How People in IT Jobs Use AI at Work (and Job Seekers Don't)

People who already have IT jobs use AI as a fast assistant for work they already understand: drafting a ticket reply, explaining an error log, writing a first version of test cases or a SQL query, then checking and fixing the output themselves. Most job seekers use AI the other way around, to produce things they can't explain, like a whole resume, a cover letter or a portfolio project. Employers hire for the first habit and filter out the second, so the skill to build is "use AI, then verify and own the result."

I'm not anti-AI. I use it every day. But when I review applicants, AI use is usually obvious, and it rarely helps the candidate. The problem isn't the tool. It's what they skipped.

How do people in IT jobs actually use AI?

They use it for speed on tasks they could do slowly without it. That's the whole difference. The human is still the one who knows whether the answer is right.

Some real, everyday patterns by role:

• IT support and helpdesk: pasting an error message (with company names and personal data removed) to get likely causes, drafting a polite ticket update for a frustrated user, turning messy notes into a knowledge base article.
• QA testers: generating a first list of test cases for a login page, then adding the edge cases the AI missed, like what happens when the OTP expires mid-flow or the user's on a slow mobile connection.
• Data analysts: asking for a first draft of a SQL query or a Google Sheets formula, then running it on real data and checking the numbers against a manual count.
• Junior developers: using GitHub Copilot or ChatGPT to explain an unfamiliar codebase, write boilerplate, or suggest why a test is failing. Then reading every line before committing.
• Everyone: rewriting a rushed Slack message to a foreign client so it sounds clear and calm.

Notice the pattern. AI produces a draft. A person checks it, fixes it, and is accountable for it. Nobody at work gets to say "ChatGPT said so" when production breaks.

How do most job seekers use AI, and why does it backfire?

Most job seekers use AI to replace the thinking, and it shows. Here's what that looks like from the hiring side:

• A resume with lines like "Spearheaded cross-functional initiatives to drive operational excellence" from someone who was a call center agent for two years. It sounds impressive and says nothing.
• A cover letter that could be sent to any company on JobStreet, because it was.
• A portfolio project the candidate can't explain. Ask "why did you use this library?" and there's silence.
• Interview answers memorized from a chatbot, delivered word for word, that fall apart at the first follow-up question.

The fix isn't to stop using AI. It's to use it the way employed people do. If you're struggling to get callbacks at all, our post on why you're not getting interviews covers the other usual causes.

What does good AI use look like in a job application?

Good AI use leaves your own facts and voice intact and makes the writing cleaner. Compare these two resume lines for a BPO agent moving into IT support.

Weak, AI-generated:

Strong, AI-assisted (you gave it the facts, it helped with wording):

The second line only works if the numbers are yours and true. AI can't know how many calls you took. You have to tell it. Our IT resume for career shifters post has more before-and-after examples.

A simple workflow that works:

1. Write your rough draft yourself, ugly is fine, with real facts and numbers.
2. Ask the AI to tighten it: "Make this clearer and shorter. Don't add anything that isn't in my draft."
3. Read the output and delete anything you didn't actually do.
4. Read it out loud. If it doesn't sound like you, fix it until it does.
5. For a project, write down how you'd explain each part to an interviewer before you call it finished.

How do you show AI skills to employers without faking it?

Show a real task where you used AI and checked the result. That's the AI skill employers actually want, and almost nobody demonstrates it.

Ideas you can build this month:

• IT support: a short knowledge base of five troubleshooting guides for common issues (printer offline, VPN won't connect, Outlook not syncing). Note in the README that you drafted with AI and tested each step on your own machine.
• QA: a test case sheet for a real public website, where you list which cases the AI suggested and which edge cases you added. That second column is the interesting part.
• Data: a small Google Sheets analysis using public data, with your SQL or formulas and a note on how you verified the totals.
• Junior dev: a small app with a README section titled "What I used AI for and what I wrote myself."

In interviews, a line like this lands well: "I used ChatGPT to draft the test cases, then I found three it missed, mostly around expired sessions. Those three are the ones that would have reached users."

Doing this alone is possible. What usually goes missing is someone pushing back on your explanation before an interviewer does. In the Career Shifter Track, the project you build gets reviewed by a coach, and you explain it out loud in mock interviews, so the follow-up questions don't surprise you.

What should you never paste into AI tools at work?

Never paste confidential data into a public AI tool unless your company explicitly allows it. In the Philippines, many IT and BPO employers handle client data covered by the Data Privacy Act and strict client contracts, and pasting it into a chatbot can get you fired on the spot.

Keep these out:

• Customer names, phone numbers, account numbers, addresses
• Passwords, API keys, server addresses, VPN configs
• Internal source code, unless your company approves a specific tool
• Screenshots of client dashboards or tickets with personal details

When in doubt, ask your team lead which tools are approved. Many companies now provide an internal or enterprise version for this reason. Asking that question in your first week also makes you look careful, which is a good look.

Will AI replace entry-level IT jobs?

AI is changing what entry-level IT work looks like more than it's deleting the jobs outright, at least for now. Tasks like writing a first draft of documentation or simple test cases are faster, so employers expect juniors to do more with that time and to catch what the tool gets wrong. The people most at risk are the ones whose only skill is producing output they can't check.

If you're worried the field is full, our honest take on whether IT is oversaturated in the Philippines is worth a read. And if you haven't picked a role yet, start with the free IT Job Starter Kit, then try the Career Shifter preview lessons to see how a project gets built step by step.

Frequently asked questions

Is it okay to use ChatGPT to write my resume?

Yes, as an editor, not an author. Write the draft with your real facts first, ask the AI to tighten it without adding anything, then remove any line you can't back up in an interview.

Do IT companies in the Philippines allow AI tools at work?

It depends on the company and the client. Some provide approved enterprise tools, others block public chatbots because of client contracts and data privacy rules, so ask your team lead before using any tool with work data.

What AI skills do employers look for in entry-level IT hires?

They look for people who can use AI to work faster and still check the output. Being able to explain what you asked the tool, what it got wrong and how you fixed it is more convincing than listing "ChatGPT" under skills.

Can I use AI to build my portfolio project?

You can, as long as you understand every part and can explain it. Add a short README note on what you used AI for, because interviewers will ask follow-up questions that a copied project can't survive.

Will interviewers know if my answers are from AI?

Often, yes. Memorized chatbot answers sound generic and collapse at the first follow-up question, so practice answering in your own words with real examples from your work or projects.