Just a brief look at the Tech headlines paints an AI-dominated picture. Google is launching new AI features left and right. Business schools are expanding courses to include AI-related content. Taco Bell, Pizza Hut, and KFC are even getting involved with AI too! Combined with Elon Musk’s recent prediction that “AI will be smarter than the smartest human next year,” we thought it was a good time to pause and reflect—because, yes, AI might be “smart,” but there are several important things AI will never be able to do.
No matter how advanced it becomes, AI will always have its limitations. Even if we find ourselves in a Taco Bell run by robots someday, those bots won’t be able to compete when it comes to empathy, critical thinking, and self-monitoring. For those of us in the business of people and talent acquisition, this is something we need to hold on to in the years to come.
According to Josh Bursin, HR and thought leader, AI isn’t something we need to be afraid of. Rather, AI is a tool we need to learn about and prepare for. He knows as well as we do that there are benefits to weaving AI into our recruiting processes and day-to-day routines. Plug in a few requests to ChatGPT and you’ve quickly got a job description. Use the right tool, and you’ve got applications summarized in a snap. Chatbots, automated messaging, and language optimization stand to usher candidates through your hiring process seamlessly.
Do these sound like processes that should scare hiring teams? Not at all. They’re wins for all of us! That said, we still want to approach AI thoughtfully. Without the continued involvement of people, we risk doing what we should be afraid of…taking the “human” out of human resources.
It might feel like AI is taking over the world, but AI simply cannot do everything. Read on for the most significant limitations TA and HR teams should be aware of.
AI doesn’t have human emotion or the ability to empathize. So, when we apply AI to our HR processes, it’s not a replacement for true relationship building. AI can’t share the ins and outs of your company culture, collaborate with other decision-makers, or relate to the human experiences that influence mood or personality. Making person-to-person connections and building strong teams will remain the responsibility of HR professionals.
If you find yourself thinking, “Finally, we can get rid of feelings in HR!,” consider this: even if holding people at arm’s length makes your job seem easier, it’s bad for business in the long run. When we allow space for authenticity and empathy at work, it improves innovation, team performance, and retention.
To take this a step further—AI might be able to summarize the key points of a resume, but it won’t know who will enhance team dynamics. And while hiring professionals shouldn’t rely on intuition to make decisions, intuition can be a signal to pay closer attention. AI platforms aren’t able to pick up on those subtleties.
You might remember a grade-school math teacher telling you to “show your work”, but showing work is another skill that AI doesn’t have. AI might share the answer to a problem, but we don’t always know how it got there. After it’s programmed, the rest of AI’s decision-making process can get lost in an “AI black box.” As such, AI could reject candidates or make hiring suggestions without good evidence.
AI can’t make thoughtful connections either. For example, if AI screened a resume and saw that an applicant worked at their last job for only six months, that’s the information it’d offer. It’d take an attentive recruiter to learn that the applicant had to leave work to overcome an illness, care for a family member, or pursue travel—all experiences that help shape a well-rounded candidate.
Similarly, if a job seeker worked at a start-up that saw rapid growth, AI might be able to explain timelines and the company’s net worth. It’d take a recruiter to infer that the applicant is likely flexible, self-starting, and a creative problem-solver.
AI can’t fact-check itself. If the data model AI is referring to is incorrect, biased, or incomplete, whatever it produces will be flawed. If you’re using AI to research hiring statistics but it was based on an incorrect or incomplete data model, you could be making decisions based on fiction rather than fact. If you’re screening candidates based on an algorithm that favors male applicants or recent graduates, you’re going to have an imbalanced talent pool. Amazon had this very problem back in 2015.
It’s also important to remember that AI doesn’t have a system of ethics or morals. Ensuring data security, accuracy, and fairness remains the responsibility of users. For instance, if hiring teams are letting algorithms make final decisions without any human oversight, AI isn’t wise enough to say “This is concerning.”
It’s still up to people to raise the alarm. If users notice inappropriate use of protected information, flawed results, or instances of plagiarism, it’s up to them to regulate and correct these issues too.
Companies using AI must understand both the strengths and weaknesses of AI. Especially in the ever-evolving field of talent acquisition, it’s up to users to take advantage of AI’s benefits while remaining aware of its shortcomings. Most importantly, HR professionals should know that as advanced as AI gets, it’s never a replacement for true human insight and connection.
As you weave AI into your hiring process, use vendors that consistently ensure the ethical treatment of people, will maintain data security, and support, rather than overtake, your decision-making process. Connect with an expert at Clovers to learn how we can help you achieve those goals today.
Scot is a successful HR technology entrepreneur and advocate for conscious inclusion. Passionate about helping others succeed, he’s committed to improving the hiring process for employers and job-seekers every step of the way.