Application rejected in minutes? It's not you, it's the algorithm

November 13, 2026


<p>Application rejected in minutes? It's not you, it's the algorithm</p>

Last updated November 13, 2026 

 

Picture someone who has just been made redundant after 15 years with the one organisation. They are good at what they do. They update their CV, apply for 40 roles in a month and hear back from almost none. Some of the rejections land within minutes.

 

It is easy to blame the market, or to start doubting yourself. But increasingly, the first reader of a CV is not a person at all.

 

AI now sits on both sides of hiring. Candidates use it to write and tailor their applications. Employers use it to sort, score and screen what comes back. Around 62% of Australian organisations used AI moderately or extensively in recruitment in 2024, according to the Australian Responsible AI Index.

 

So both sides are using similar technology, and both sides are frustrated. Employers say they cannot find the right people. Candidates say they apply and hear nothing. I hear both sides of that conversation every week, and the gap between them is often smaller, and more fixable, than people think.

 

What the research tells us

 

The most telling finding comes from employers themselves. In Harvard Business School and Accenture's Hidden Workers research, 88% of employers surveyed in the US, UK and Germany agreed that qualified candidates are filtered out simply because their applications don't match the exact wording of the job description. Lead researcher Joseph Fuller argues that the push for efficient hiring is creating much of the talent shortage employers complain about.

 

HR leaders here are uneasy too. Research by AHRI and QUT found that 39% of HR leaders using AI in recruitment believe it discriminates. Yet fewer than one in four organisations had actually checked their tools, and almost half of those that did found a problem.

 

Dr Natalie Sheard at the University of Melbourne found that AI hiring systems may enable, reinforce and amplify discrimination against historically marginalised groups. Australia still has no specific laws governing how these tools work. That leaves the responsibility with the employer. But from next month, employers will at least have to be open about using them.

 

Overseas, the legal ground is shifting. In the United States, Derek Mobley claims he has been passed over for more than 100 roles since 2017 at organisations using the Workday platform. On 22 June 2026, Judge Rita F. Lin ruled that the core discrimination claims can proceed. Nothing has been proven yet, but it is a clear signal of where scrutiny is heading.

 

What changes on 10 December

 

From 10 December 2026, organisations covered by the Privacy Act must say in their privacy policy when a computer program makes, or substantially helps make, a decision that could significantly affect someone, using that person's personal information. Hiring is squarely in that territory. Deciding whether a candidate moves forward or is rejected clearly affects their interests.

The policy needs to set out the kinds of personal information used and the kinds of decisions involved. The Office of the Australian Information Commissioner released its final guidance on 30 September, and it is broader than many employers expect:

 

  • A person in the loop doesn't take a tool out of scope. If the software's output is a key factor in a decision, it counts, even when a person makes the final call.
  • The onus sits with the employer, not the vendor. If your applicant tracking system screens or scores candidates, you are the one who has to disclose it.
  • It covers tools you already have. The rule applies to decisions made from 10 December, no matter when the system was put in.

 

It's worth being clear about the limits. This is a transparency rule. It doesn't give candidates a right to challenge a decision, and it doesn't require employers to tell each applicant directly.

Even so, it changes the conversation. To write an honest privacy policy, HR has to know what its tools actually do. For many organisations, that will be the first time anyone has looked.

 

Where it goes wrong

 

The problems are rarely dramatic. Usually it's a small setting doing exactly what it was told to do.

 

  • A career break becomes a red flag. Sheard found screening systems using gaps in employment history to rule people out. Women are more likely to step out of work to care for family, so a gap quietly stands in for gender. Returning parents, carers and people who have been through redundancy are hit hardest.
  • Assessments measure the wrong thing. A career coach in Sheard's study described a neurodivergent client who topped his university course and still couldn't get through personality assessments.
  • The tool learned somewhere else. For HireVue, a widely used video interview platform, about 6% of applicant data comes from Australia and New Zealand, while close to 80% comes from North America. Australian accents and career paths may simply be read less accurately.
  • The machine learns our old habits. Amazon's experimental system taught itself to downgrade CVs that contained the word “women”, because it was trained on a male-dominated pool of past applicants. It was scrapped before it was used in live hiring, but the lesson stands.
  • Formatting gets in the way. A strong CV can be ruled out because the system couldn't read a skills section set in two columns.
  • No one is home. In the Workday case, applicants say they were declined within minutes of applying. That timing suggests no person was involved.

 

If you're looking for work: make AI work for you

 

AI can be a real help in a job search. It's a good drafting partner, a quick research tool and a fast way to tailor an application. It works best when it helps you say what is already true about you, in the words the employer is using.

 

  • Use their words. If the ad says “stakeholder engagement”, don't write “relationship management”. Use the actual job title and the key skills listed, naturally and honestly.
  • Treat AI as a first draft. Recruiters spot generic AI copy quickly. Let it structure and tailor, then rewrite it so it sounds like you, with real examples and real numbers.
  • Keep your CV simple. One column, standard headings (Experience, Education, Skills), no tables, text boxes or graphics. Save it as Word or a plain PDF.
  • Explain a break, don't hide it. One line does the job: “Career break: full-time carer” or “Career transition and upskilling”.
  • Read the knockout questions twice. Questions about salary, location, visa status or years of experience often trigger an automatic no. This is where good applications quietly end.
  • Don't rely on the portal alone. Find the hiring manager or recruiter on LinkedIn, send a short note and ask for referrals. A warm introduction goes around the filter entirely.
  • Ask for adjustments. If you're neurodivergent or have a disability, you're entitled to ask for an alternative to a timed test or video interview.

 

There's some good news on the way, too. From 10 December, organisations must say in their privacy policy if software helps decide who moves forward, and what information it uses. Before you apply, it's worth a quick look. If a system is involved, you'll know to write for it, and to find a real person to talk to as well.

 

Most of all, please don't take the silence personally. Being screened out by a system is not a verdict on your ability. Change how you apply, not who you are.

 

If you're hiring: check the settings, not just the software

 

Most AI recruitment tools are off-the-shelf products. The same applicant tracking system or screening engine is sold to a mining company, a hospital and a software start-up. What makes it work for your organisation is how it's set up.

 

Those settings are usually decided once, during implementation, often by a project team racing to a go-live date. Then they are left alone for years. The keyword list, the knockout rules, the score thresholds and the ranking weights quietly decide who you ever get to see. If your shortlists look thin, the cause may be a setting rather than the market.

 

A practical review looks like this:

 

  • Separate hard knockouts from ranking signals. Very few criteria genuinely need to be a hard no. Everything else should lower a score, not delete a person.
  • Read a sample of rejected applications. If you find people you would have interviewed, your filters are too tight.
  • Watch for proxies. Employment gaps, graduation year, continuous service, postcode and photo-based scoring can all stand in for age, gender, caring responsibilities or background.
  • Ask where the tool was trained. Find out how much of the vendor's data comes from Australia and New Zealand, and what local testing has been done.
  • Look hard at the job ad. A long list of “essential” requirements becomes a long list of automatic rejections.
  • Keep a person involved at rejection, especially at volume, and especially for internal applicants.
  • Be open with candidates. From 10 December, your privacy policy must say where automated tools help make hiring decisions. Go further than the law asks: tell candidates directly when AI is part of the process, and give them a way to ask for an adjustment.
  • Measure who you missed. Time to hire and cost per hire say nothing about the talent you filtered out.

 

The technology isn't the villain

 

It's worth saying clearly: used well, these tools can be fairer than people. Professor Andreas Leibbrandt at Monash University found that recruiters scored women lower when gender was visible, and that the gap closed when it was hidden. The problem is how a system is set up, and whether anyone checks the results.

 

Right now, the pressure is building on both sides. Application numbers have climbed sharply, partly because candidates use AI to apply for dozens of roles at once. Employers respond by tightening their filters, which catches even more good people. Each side is reacting to the other, and the person in the middle is the one who pays for it.

 

Job seekers can't break that loop on their own. It takes a deliberate decision by employers.

 

How Career Money Life helps

 

When roles are made redundant, this is the job market your people step into. Outplacement built for a pre-AI market leaves them underprepared for it.

 

Our career transition programs are built around choice. Instead of a fixed package, each person receives marketplace credits and chooses the support that fits their situation, their industry and their stage of life. For someone facing AI screening, that often includes:

 

  • AI-ready CV and LinkedIn support, with CVs written to be read cleanly by screening tools and tailored to the language of each ad
  • Practical AI coaching, so people use these tools to tailor and research without sounding generic
  • Interview and assessment preparation, including video interviews and psychometric or personality assessments
  • A targeted search strategy built on referrals, direct approaches and networks rather than portal volume
  • Career coaching, which matters most when silence is the only feedback someone is getting
  • Adjustment and accessibility support for people who need an alternative to a standard assessment

 

CareerCanvas assessment sits underneath every program. It helps each person understand their strengths, values and direction before they start applying. Support is delivered by a curated network of Australian providers, and HR sees engagement and outcomes in real time, so you know what your people used and where they landed.

 

If you'd like to talk about preparing your people for the market as it really is, we'd love to hear from you at careermoneylife.com.

 

About the Author

 

Sandy Hutchison | CEO and Founder, Career Money Life

Sandy is a leading Australian HR professional and thought leader on career transition, coaching, and organisational change. As the founder of Career Money Life, she has dedicated her career to humanising the redundancy process, giving individuals the agency, structure, and choice they need to move forward with confidence, care, and purpose. She is a specialist in the people pillar of organisational change, with deep expertise in wellbeing, coaching, and marketplace-driven transition solutions.

linkedin.com/in/sandyhutchison

 

Return to Blogs

Partner with us in supporting your people's key career moments

Book a Meeting