In the fast-paced world of Human Resources (HR), efficiency and effectiveness are two guiding principles that underpin success. As an HR practitioner, you’re likely juggling multiple responsibilities at once and rarely enough time to get things done; and done well. With such demanding roles, wouldn’t it be fantastic if you had a digital assistant that could expedite some of your routine tasks while maintaining high quality? Enter ChatGPT.
The Power of AI in HR
ChatGPT is an advanced AI developed by OpenAI, designed to generate human-like text. It has been trained on diverse internet text but can be fine-tuned to tasks such as drafting emails, writing reports, and creating content. It’s just as adept at assisting HR practitioners in their routine tasks, helping to draft job descriptions, generate interview questions, and create evaluation rubrics in a matter of seconds. In this blog, I’m going to discuss interview question generation but before delving into this I need to mention a few things about using any AI tools.
Important Things to Know Before Using ChatGPT
There are a number of things to be aware of before embarking on employing ChatGPT as your personal HR assistant.
First, never, ever, use PIIs in a prompt for ChatGPT. OpenAI cautions against this because they cannot guarantee confidentiality and security. If need be, use pseudo placeholders for such information. Instead of a person’s real name, use Jane Doe or John Doe. Instead of your organization’s real name, use XYZ Corp.
Second, remember that ChatGPT is designed to be a super chat bot, it is optimized to emulate conversation between humans. To that end, ChatGPT, in my estimation, passes the Turing Test. In 1950, Alan Turing, a British mathematician, posited the Turing Test of a machine’s ability to exhibit intelligent behavior. Turing asserted that one could say a machine passed the test if its conversational responses are “equivalent to, or indistinguishable from, that of a human.” Keep in mind that the measure for passing this test is to sound like a human. The measure is not the accuracy of the response. Humans are fallible. Therefore a machine trained to sound like a human can also be fallible. It is important to always review a ChatGPT response critically to detect errors.
Finally, working with ChatGPT is an iterative process. You can prompt ChatGPT with a general question or just tell it to do a general task. But it does not have to stop there. Once ChatGPT has provided a response to your initial prompt, you can prompt it further by adding additional detail and telling it to revise its response. This is the real utility of ChatGPT. Again, it’s designed to be conversational; so converse with it. This process of iterative prompts is called prompt engineering. The better you become at prompt engineering, the better the responses from ChatGPT.
Crafting High-Quality Interview Materials with ChatGPT
Here’s how you could use ChatGPT to help in evaluating a candidate for a job. In this example I’m using a Data Scientist position to demonstrate how ChatGPT can be used by an HR practitioner in a situation where the HR practitioner may not be well versed in the nature of occupation. Begin by prompting ChatGPT with the following:
“Generate a set of interview questions for candidates for a Data Scientist position that requires expertise in Python, machine learning, and data visualization where the job description is the following [paste job duties]”
In a matter of seconds, ChatGPT might generate questions such as:
1. “Can you describe a project where you leveraged Python for data analysis? What were the outcomes?”
2. “Can you explain a situation where you implemented machine learning algorithms to solve a complex problem?”
3. “Describe your approach to creating an effective data visualization. Can you provide an example?”
Once ChatGPT has produced a set of questions, you could then prompt ChatGPT to create an evaluation rubric for each question. For example:
“Create an evaluation rubric for the question, ‘Can you describe a project where you leveraged Python for data analysis? What were the outcomes?’ with a weak and a strong response.”
ChatGPT could then generate a rubric as follows:
Question: Can you describe a project where you leveraged Python for data analysis? What were the outcomes?
Weak Response: The candidate provides a vague description of the project, lacks specifics on their role and the techniques used, and doesn’t clearly articulate the project’s outcomes.
Strong Response: The candidate provides a detailed description of the project, including the problem, the dataset, and the specific Python libraries and techniques used. They explain their role in the project and provide quantifiable outcomes, demonstrating their ability to use Python for data analysis effectively.
Efficiency at Your Fingertips
The main advantage of using ChatGPT is the speed and efficiency with which you can produce high-quality material particularly where you may not be conversant in the occupation. Instead of spending long hours crafting interview questions and responses from scratch, an HR practitioner can use ChatGPT as a starting point, generating a framework literally in seconds, which can then be refined and customized.
In essence, ChatGPT allows HR practitioners to allocate their time more effectively. By handling the preliminary drafting work, the tool frees up HR practitioners to spend more time collaborating with hiring managers, refining the material, and tailoring the hiring process to suit the specific needs of the organization and role.
Conclusion
In today’s fast-evolving HR landscape, leveraging AI tools like ChatGPT is not just a value-add; it’s a competitive necessity. By using ChatGPT to assist in generating interview questions and rubrics, HR professionals can enhance their efficiency, streamline their processes, and ultimately, make better hiring decisions. Harness the power of AI and transform your HR practices today. AI is a tsunami. You can either be riding the wave – in a safe and responsible manner – or you can be on the beach inundated and helpless.