AI in Hiring: Can We Trust Technology to Hire Responsibly?

The Promise of AI in Hiring

Let’s start with what machines actually get right.

Recruitment has always been a seriously time-consuming process. If you’ve ever been on the hiring side, you know exactly what I mean. Hundreds (sometimes thousands) of resumes to sort through, endless interview scheduling, assessment reviews, follow-ups… it’s exhausting. No surprise that HR teams often feel like they’re drowning.

And this is where AI really shines.

When it’s used properly, AI can take a huge amount of weight off recruiters’ shoulders. For example, it can:
• Automate the boring stuff: things like initial CV screening, sending reminders, or handling routine emails.
• Boost efficiency: instead of spending hours sifting through resumes, it can process thousands in seconds.
• Spot patterns we might miss: like the specific skills or behaviours that tend to predict success in a role.
• Improve candidate matching: using data to recommend roles that might fit someone even better than what they applied for.

And we’re already seeing this play out. In Singapore, more companies are adopting AI-driven recruitment tools, especially in high-demand sectors like tech, energy, and data centers. With talent shortages and fierce competition, employers are turning to AI to help them make faster, more informed decisions.

The Hidden Biases Inside the Machine

People often forget that for AI: it’s only as objective as the data we feed it.

If the historical data is full of human bias, favoring certain genders, races, schools, or backgrounds, then the AI will simply learn those patterns and repeat them. In fact, it can even amplify them, because machines don’t question their instructions the way we do.

Let me give you a well-known real-world example.

Amazon built an AI tool to help them spot top candidates for technical roles. Sounds great in theory, right? But after months of testing, the engineers found something deeply concerning: the system kept downgrading any CV that contained the word “women’s.” For example, “women’s chess club captain” or “leader of the women’s coding group.”

The AI wasn’t “anti-women.” It was learning from years of hiring data in a male-dominated environment. In other words, the machine simply copied past patterns — and those patterns weren’t fair. The company eventually scrapped the tool, but the lesson was crystal clear: if we feed AI biased data, it will produce biased decisions.

And this isn’t an isolated incident.

Multiple studies in the US and UK found that AI hiring systems unintentionally favoured candidates with “Caucasian-sounding” names, or downgraded applicants from less well-known universities. The systems weren’t designed to discriminate against. Instead, they simply inherited the skewed data they were trained on. This issue is closer to home than people think.

In Singapore, the Tripartite Alliance for Fair and Progressive Employment Practices (TAFEP) has already warned employers that relying on untested or opaque AI tools could breach fair hiring guidelines.

And let’s be honest. The consequences go far beyond a bad headline. Unfair hiring practices can lead to reputational damage, regulatory penalties, and the loss of great talent who never even get a chance to be seen.

Beyond Bias: The Black Box Problem

Bias isn’t the only thing we need to worry about. There’s another big challenge in AI hiring: what people often call the black box problem.

A lot of AI hiring tools run on deep learning models or proprietary algorithms that are incredibly complex. Sometimes even the people who built them can’t fully explain why the model made a certain decision. So when someone asks, “Why was this candidate rejected?” the answer ends up being something vague like: “That’s what the model decided.”

And honestly? That’s just not good enough.

Hiring isn’t a casual process. These decisions affect real people’s careers, income, and futures. Candidates should be able to understand how decisions about them are made. Employers need to be able to explain those decisions clearly, especially if regulators or applicants challenge them.

This is exactly why responsible AI frameworks around the world are putting such a big emphasis on explainability.

Here in Singapore, initiatives like AI Verify encourage companies to test and validate how their AI systems behave. Over in Europe, the EU AI Act sets strict standards for transparency and accountability, especially for high-risk AI systems like those used in hiring. If AI is going to help us make decisions about people, then those decisions must be understandable, defendable, and accountable.

Responsible AI: A Partnership, Not a Replacement

When we talk about responsible AI in hiring, here’s the golden rule: AI should be your assistant, never your authority. The whole point is to give recruiters more time for the important stuff, actually talking to people, understanding what motivates them, spotting potential, and building real human relationships.

One of the best ways to keep this balance is by using what’s called a Human-in-the-Loop (HITL) approach. And don’t worry, it’s not as technical as it sounds. HITL simply means that a human always has the final say.

If an algorithm ranks someone poorly, a recruiter still looks at the application before taking action. If an AI chatbot screens candidates, it sends the “not sure about this one” cases to a real person. You keep the efficiency of AI, but you never hand over the steering wheel completely.

And believe me, this matters more than people realise.

Let me share a story that perfectly captures why.

Not long ago, a manager in the tech industry shared something wild on Reddit. A job seeker had posted about getting a rejection email seconds after applying. Apply at 10:56, rejection at 10:56. The manager recognized that pattern all too well, because his team had been struggling for months to find a single qualified candidate… which didn’t make sense.

So, he decided to run a little experiment. He created a new email, tweaked his own CV, changed the name, and applied to his own open role. Instant rejection. Not even a second glance. That’s when it clicked: nobody was reviewing anything. To make things worse, HR kept reporting that candidates “didn’t pass initial screening,” when in reality, nobody passed because the system simply blocked everyone.

This is exactly why HITL is essential.

AI can filter, summarize, and prioritize. But it can’t replace human judgement. Without oversight, even the best tools can become silent bottlenecks that cost organizations great talent and erode trust in the hiring process.

The Global Picture: What’s Coming Next

All around the world, governments and companies are wrestling with the same big question: How do we make AI fair, transparent, and accountable, especially when it’s used in hiring?

Different regions are taking different approaches, but the message is consistent.

In Europe, the EU AI Act put recruitment tools in the “high-risk” category. That means strict rules, mandatory safeguards, and clear human oversight. No fully automated hiring allowed.

In the US, the Equal Employment Opportunity Commission (EEOC) are actively investigating cases where algorithms might be discriminating against candidates. They’re treating AI bias the same way they would treat human bias.

Across Asia, countries like Japan and South Korea are doubling down on transparent AI governance. They know that if AI is going to drive their economies, it has to be trusted, both locally and globally.

And then there’s Singapore. With its AI Verify program and pro-innovation stance, Singapore has positioned itself right in the middle of this global conversation. It’s becoming a bridge between ethical leadership and technological progress, showing that you can encourage innovation and uphold strong values at the same time.

Trust Is Earned, Not Automated

So, can we trust machines to hire? Honestly, it’s not a simple yes or no.

At the end of the day, AI isn’t here to replace the human side of hiring, it’s here to strengthen it. When we use AI responsibly, with human judgement guiding every important decision, we get the best of both worlds: smarter processes, fairer outcomes, and a hiring journey that actually respects the people in it.

Machines can help us work faster and see patterns we might miss, but it’s humans who bring empathy, context, and values, and that’s what truly defines a great hire. The future of recruitment isn’t about choosing between AI or humans. It’s about using AI with humans, working together to build workplaces that are fair, transparent, and genuinely people-first.