What AI Can’t See: The Growing Disconnect in Hiring

July 11, 2025

By Scott Michajluk | Fractional COO | Operational Growth Advisor

I've spent over two decades leading operations, building teams, and helping businesses grow responsibly. In that time, I've seen hiring evolve from handshake deals and gut instinct to applicant tracking systems, keyword filters, and now AI-powered decision engines. But here in mid-2025, we’re past the point of novelty.

In recent months, I’ve heard a familiar refrain from business owners and candidates alike: “The system feels off.”

On paper, the job market appears steady. And while for the past year reports continue to cite that unemployment remains fairly low, you can't help but wonder if those numbers are based in reality. A daily scroll through your LinkedIn feed, and a different picture unfolds. Candidates from senior leaders to fresh college grads applying for roles without acknowledgment. Hiring teams overwhelmed by hundreds if not thousands of applicants for a single role. Talented professionals questioning their value, and not because of performance, but because of silence.

As someone who has spent a majority of his career building operations teams and advising founders through periods of scale, I’ve come to believe this issue goes far deeper than headlines or metrics. It reflects a fundamental shift in both how hiring is being conducted and how candidates are applying. Leaving us all to wonder if AI is now hindering more than aiding the hiring process.

Have we outsourced too much?

Artificial intelligence is not inherently problematic. Used well, it can help identify patterns, flag inconsistencies, and surface resumes that deserve a closer look. But when it comes to hiring, has the AI pendulum swung too far?

When it comes to hiring, AI should be a tool to support human decision-making, not replace it. We see qualified candidates being removed from consideration before a human ever sees their name. Application systems default to rigid criteria, scoring algorithms, and formatting preferences. The cost is substantial because we are filtering out human nuance in pursuit of operational ease.

This is not a knock to technology. But when the tools begin to obscure the very thing we’re trying to assess—accountability, judgment, leadership, capability—we ought to pause and ask whether efficiency has become the enemy of effectiveness.

What employers can do differently

Hiring managers and business owners are under pressure. Some roles receive over a thousand applications. Time is limited. Expectations are high. But that reality does not absolve us of the responsibility to treat candidates with respect or approach the process with care.

If you are hiring, here are a few practices worth considering:

• Refine your filters, but read beyond them. AI may help surface likely matches, but it should never be the final word. Make time to review a portion of applications that fall outside the parameters. You will find value where the model did not.

• Acknowledge the application. Even a short, respectful message can restore a sense of dignity to the process. Silence may be efficient, but it is not professional.

• Choose clarity over cleverness in your job descriptions. When requirements are ambiguous or inflated, you attract confusion, not talent.

• Evaluate the human, not just the resume. We need to put Human back into Hiring. A resume cannot reveal how a candidate mentors others, navigates conflict, or improves systems. These qualities matter, and they are best judged through conversation.

These adjustments do not require a complete overhaul, just a willingness to reintroduce discernment into the process. They are small corrections, but they restore the balance that many organizations have lost.

A word to candidates

For those on the other side of the table, I know this is a difficult season. I see amazingly talented executives and professionals second-guessing their value. I see early-career candidates trying to reverse-engineer “the perfect application.” And I see too many people trying to write for a machine instead of representing who they actually are.

Some constructive feedback:

• Stop trying to be perfect on paper. AI-driven filters reward conformity, not competence. If you write solely to appease the algorithm, you erase the very qualities that make you distinct. Use your voice, not one generated by an AI prompt.

• Apply where there is alignment, not just availability. A targeted, well-articulated application to a company you admire carries more weight than a scattershot approach to thirty positions.

• NETWORK, NETWORK, NETWORK. With the current state of the job market and hundreds if not thousands of applicants vying for a single role, this maybe the post important piece of feedback to consider. Create that human connection, reach out, be proactive, let your tone reflect your actual working style. Machines may prefer keywords, but people hire people.

No system no matter how sophisticated can truly measure potential, character, or judgment. Those qualities remain human and must be assessed by humans.

The bigger issue

I believe hiring is fundamentally about trust. A company is trusting someone with its time, resources, and reputation. A candidate is trusting an organization with their livelihood and future. When that exchange becomes impersonal, both sides suffer.

In today’s environment, we need to regain that trust by reintroducing the human element. That does not mean rejecting technology outright. It means finding the right balance and ensuring that it does not overtake the responsibility of discernment.

The volume may be high, but the obligation to lead with empathy remains.

Companies want to build strong teams. They want to find the right leaders and place the right people into the right roles. It never hurts to take a step back and do a reset. Maybe slow down just enough to hire wisely. Structure operations so that people are not lost in the process, but rather seen, evaluated, and brought in with purpose. It’s not a radical approach. It’s just responsible leadership.

And perhaps, right now, that’s what we need most.