In early 2023, ChatGPT set the record for the fastest-growing consumer app, counting 100 million monthly active users a mere two months after its release.
Its impact was equally significant on the business front. This is true even in talent acquisition, where talent teams were using AI recruiting tools long before the interest in generative AI.
But despite the widespread presence of AI tools, there is still confusion about what exactly these recruiting solutions can accomplish and how. And when we don’t clear this confusion, we create space for the misuse of AI recruiting software, which can lead to costly consequences.
In this guide, we’ll cover all of the essential aspects of AI in recruiting to help you leverage these tools in an informed way.
What is AI?
Artificial intelligence (AI) is a technology that uses problem-solving and decision-making capabilities to solve different tasks and processes. It’s able to do so at a scale and speed that human beings are unable to achieve.
This technology can perceive information, learn, solve problems, and generate visual, written, or audio content. AI-supported solutions are widely present in both business and consumer-facing environments. For example, speaking to Siri on your phone requires AI, as does using a customer service chatbot.
Many business leaders see AI as an opportunity to offload time-consuming tasks so that people can focus on higher-value decision-making. They’re also keen to leverage the potential of AI in analytics where this technology can extract valuable insights from large volumes of data.
What is machine learning?
The terms machine learning and AI are often used interchangeably, but they are not the same. Machine learning is a branch of AI that uses algorithms and data to learn, while AI is a broad term for any kind of advanced technology imitating human cognitive processes. According to Douglas Eck, a senior research director at Google, “Most of what we see in AI today is really machine learning: endowing computer systems with the ability to learn from examples.”
In talent acquisition, machine learning can analyze vast amounts of candidate data — more than a human being is capable of. Then, it identifies trends and insights (such as candidates who had a fast career progression) that help you make more informed hiring decisions.
What is generative AI?
Generative AI refers to AI technology that can produce videos, images, texts, and other types of content. This is possible thanks to the ability of machine learning models to ingest and learn from large data sets. Then, the model “can take what it has learned from the examples it’s been shown and create something entirely new based on that information,” per Eck.
If you want to improve your candidate experience, you can use generative AI to analyze the results of feedback surveys. Then, it can summarize the results to help you understand the biggest drivers of candidate dissatisfaction.
What are large language models (LLM)?
Large language models, like ChatGPT, are a subtype of generative AI. They are able to understand and create text in a way that’s similar to a human being, as well as translate, summarize, and rewrite content. In talent acquisition, recruiters also use LLMs to craft outreach emails.
Conditions for success of AI in recruiting
Today, 4 out of 5 employers in the US are using AI-powered tools. In this group, AI enabled 65% to become more efficient in candidate sourcing. And it has expanded their teams’ reach without increasing their headcount. Seeing these numbers, any organization that doesn’t include AI in its talent acquisition process would likely consider it out of fear of missing out on AI’s benefits.
However, AI isn’t a guarantee of success on its own. Any AI recruiting tool needs good data and human oversight to be successful; otherwise, it will create more issues than it solves.
1. High-quality, high-volume data
AI needs high-quality data — and lots of it — to be effective. Feeding inaccurate or incomplete data into an AI system will produce unreliable results, whether you’re trying to find candidates or discover insights about the talent market. Good-quality data is accurate, complete, unique (without duplicates), valid, timely, consistent, and meets the business’s needs.
Data volume is another important point, as AI models often require large amounts of data to be trained for a specific use case. (For example, LLMs are trained on big data sets.)
However, AI experts like Andrew Ng have emphasized that large data sets aren’t required to solve every problem:
“In many industries where giant data sets simply don’t exist, I think the focus has to shift from big data to good data. Having 50 thoughtfully engineered examples can be sufficient to explain to the neural network what you want it to learn.” – Andrew Ng, Founder of DeepLearning.AI, Founder and CEO of Landing AI
So while you should consider volume as an important factor, especially if you’re using AI to get a 360-degree view of your talent pool’s skills and experience, it’s more important that the data supports quality outcomes.
We can use Findem’s AI talent acquisition and management platform as an example. It uses trillions of data points to analyze and produce candidate insights. But it also supports quality outcomes with explainable and auditable data. You can also securely plug in your data to minimize confirmation bias.
2. Human oversight
The excitement around AI, coupled with the impressive features of various AI tools, could trick you into assuming AI is too powerful to make mistakes. This is a false assumption – the responsible use of AI in recruiting needs human oversight.
Human oversight means that the AI isn’t making decisions on behalf of anyone on the team, but is an assistant in the decision-making process. AI shouldn’t decide which candidates get to move forward in the hiring process. Nor should it evaluate people on a recruiter’s behalf.
And while AI is capable of understanding your intent, you should incorporate it in your workflow in a way that enables you to confirm the AI has understood your intent.
AI also needs ongoing monitoring and guardrails against hallucinations to prevent you from using false information in the hiring process.
Advantages of AI recruiting solutions
With high-quality data and human oversight in place, you’re set to experience the advantages of AI in recruiting.
1. Faster candidate sourcing and hiring
AI tools source qualified candidates much faster than traditional methods because they have access to vast amounts of talent profiles and data. (Think trillions of data points.) The data doesn’t just come from resumes — AI is capable of aggregating talent data from many public sources, including social media. The more data you have, the easier it is to decide if a person is the right fit for the role.
And when you add automation into the mix, it’s possible to find and hire qualified candidates even faster.
Take the company RingCentral as an example. Before using Findem’s AI talent acquisition and management platform, its recruiters spent a lot of time digging through various sourcing channels to find suitable candidates. It made it difficult for the team to achieve hiring goals and understand which sourcing approaches worked best.
After implementing Findem’s platform and connecting it with RingCentral’s ATS, the team was able to search across multiple channels from one platform. Using hundreds of thousands of data sources, Findem’s attribute-based search also allowed RingCentral to find candidates with the precise mix of required skills and experience. This resulted in a 40% talent pipeline increase.
Accelerating candidate sourcing has additional benefits. Recruiters are less likely to burn out since they’re no longer investing most of their time into searching for talent and having little to show for it. They have more time to invest in engaging their talent pipeline and talking to their best candidates.
2. Improved candidate experience
The candidate experience can make or break the recruiting process. A recent survey found that 53% of candidates dropped out of the process due to poor communication. This is second only to the salary not matching the candidate’s expectations. The good news is that AI can improve the candidate experience with personalization, a smooth application process, and regular updates regarding their application’s status.
Some companies today use conversational AI (chatbots) to communicate with potential candidates about open positions. Instead of waiting for a recruiter’s response, the chatbot will answer any questions in real time. And if the candidate wants to apply, they can submit their application in the chat. This type of AI solution (offered by companies like Paradox) is especially beneficial for companies with high-volume hiring needs that are unable to personally respond to every inquiry.
When communicating with candidates over email, personalization is crucial to make them feel valued. If you’re recruiting for niche or executive roles, showing you understand the candidate is table stakes to even getting noticed. Emails that show you understand their interests, seniority level, or other unique aspects of their career help you stand out in their inbox.
AI aids in personalization with data-driven insights into the candidate’s skills, experience, and interests. By aggregating talent data from many public sources, you can see who your candidates are beyond their resumes.
In the case of startup ALT, the company used AI-powered email personalization to engage and hire top talent for leadership roles. Even though the startup had a two-person recruiting team, they were able to double the response to their email outreach because they had more data on their talent pool to personalize their messages. For example, they could see which candidates had seed or Series A experience. The team could also do A/B testing to see which email campaigns performed best with different roles and seniority levels.
3. Automated repetitive tasks
Tasks such as candidate screening and interview scheduling eat up valuable time that recruiters could spend engaging with their top candidates. AI can automate these time-consuming processes with little input from the talent team.
Without automation, interview scheduling involves a lot of emails back and forth to find a time that works well for everyone involved. The process frustrates candidates and requires recruiters to spend time poring over calendars and sending emails. With automation, however, a candidate can get a list of available times and schedule a meeting with a couple of clicks.
According to Becka Klauber Richter, co-founder and president of Helpr, AI allows for “more-predictable time expectations” when it comes to criminal screening rather than having clerks perform manual checks. Mike Fitzsimmons, founder and CEO of Crosschq, emphasizes the role of AI in collecting information about the candidate’s performance at previous jobs. The data helps employers make a more informed hiring decision with less time invested.
Challenges of AI in recruiting
For all of its benefits, AI recruiting solutions come with their own set of challenges. Although these challenges shouldn’t deter you from considering AI, it’s important to be aware of them before implementing AI tools.
1. Biased algorithms
If recruiting algorithms are trained on biased data, these biases will carry over into the recruiting process and cause discrimination.
The story of Amazon’s scrapped recruiting engine is one of the most famous examples of AI bias in recruiting. Originally, the company set out to build an experimental AI hiring tool that would score job seekers on a scale of one to five. Sources even alleged that the tool was supposed to select the top candidates from a pool of resumes for the company to hire, taking the hiring decision out of human hands.
However, Amazon soon discovered the AI system was discriminating against female applicants. It had been trained on resume data that came from mostly male candidates and downgraded resumes that mentioned words like “women’s.” Ultimately, the company did not use the AI tool to make any hires.
This is a prime example of a) why human oversight is necessary when using any kind of AI recruiting platform and b) why we should treat AI tools like assistants, not managers.
Instead of using AI to score applicants, we can use it to create holistic talent profiles. By combining a candidate’s publicly available data with their resume, AI helps us better understand the candidate and how their expertise fits in with the company’s needs. Most importantly, the AI isn’t making any decisions, but just providing insights.
2. Increased compliance risks
AI systems collect a lot of data, which may also include personally identifiable information (PII) and intellectual property (IP). This poses a compliance risk if your internal processes violate any applicable legislation.
New York City’s AI hiring law, for example, requires employers to audit their AI tools for bias. Additionally, they must notify job applicants that they’re using such tools. Failure to do so carries fines from $500 to $1,500 for every violation.
Compliant and responsible use of AI also means proactively implementing processes and sharing best practices that lower the risk of accidentally exposing PII and IP. For example, you should have a policy that regulates the use of LLMs.
Any AI talent acquisition tools you use should also incorporate middleware that separates the tool from public LLMs. The role of the middleware is to anonymize the inputs and prevent private data from becoming public.
3. Overworked and burned-out recruiters
We’ve come to associate words like efficiency and higher productivity with AI. But it’s entirely possible for an AI tool to produce incremental time savings — while adding to recruiters’ workloads.
Let’s say that a company implements an AI talent acquisition platform with the primary goal of finding more passive candidates. The platform allows recruiters to find 10 times more potential candidates than before, but it has limited search filters based on keywords and Boolean strings. The recruiters are unable to easily find the profiles of the most qualified candidates, so they resort to manual profile screening.
Now, the recruiting team has more work added to its plate and needs another tool to filter candidates. To avoid this, make sure that the AI solution helps you find many candidates and also offers robust search features to sift through the profiles quickly.
AI in talent acquisition: 7 use cases
From market intelligence to candidate relationship management, AI has many applications for acquiring and nurturing talent.
1. Understand your talent pool
Before you even publish a job posting, AI-enabled insights can help you understand the talent market. For example, if you’re hiring for an office-based role in Nevada, Findem allows you to see how many qualified candidates are available there. If Nevada doesn’t have enough options, then you can adjust your strategy to include other states, make the position a fully remote one, or offer a relocation package.
2. Generate job descriptions
A compelling job description is key to attracting active candidates. Generative AI, like ChatGPT, helps recruiters automate the writing portion of this task and craft a job ad that speaks to their perfect candidate. Augmented writing tools are also available to ensure you haven’t used any phrases that could exclude certain applicants or discourage them from applying.
3. Build searches with job descriptions
When you’re using an AI talent acquisition tool like Findem, you can plug in your job description and the platform will find suitable candidates based on your requirements. You also have the option to use an ideal candidate’s profile to discover people with similar skills and experience.
4. Accept job applications
Most companies have a careers page where candidates can submit a job application. With a chatbot, however, you can make it easier for website visitors to discover job openings, get answers to their questions, and apply. The chatbot can also suggest potential roles based on the candidate’s previous experience and introduce the candidate to jobs they might not have considered before.
These chatbots, though, should always be supported by humans. A recent Business Insider report revealed how chatbots can complicate the application process and make it difficult to schedule interviews. Without humans in the picture, job seekers have no other choice but to abandon the application.
5. Unify sourcing channels
Recruiters juggle multiple siloed channels, including social media, referrals, and past applicants, to source passive candidates. AI talent acquisition platforms that feature millions of enriched candidate profiles help you merge all of those channels in one place and save time.
6. Summarize and pitch candidate profiles
When a recruiter discovers a candidate who seems like a great fit for the role, they can use generative AI to pitch their profile to a hiring manager.
In the example below, Findem summarizes a candidate’s profile with a focus on the job requirements.
The recruiter can send this summary to the hiring manager. Or, they can ask Findem to write an even more personalized profile: “Can you summarize this person’s profile for a hiring manager who values fast career growth?”
These AI capabilities can save recruiters hours of time that they can reinvest into the candidate experience.
7. Engage candidates
Regular, personalized communication is important when you’re actively hiring for a role. But it also matters when you’re nurturing your talent pipeline for future roles.
It improves your employer brand and shows you see your applicants as people rather than just an email list. With AI talent insights, you can segment your applicants by industry, role, seniority, etc.
Generative AI can then craft an email with a personalized tone and length. It can even tailor the copy to the type of email, whether it’s the first email in a sequence or a follow up.
Discover the precise talent you need 50% faster
Although recruiting AI tools are advanced, AI will never replace people. Instead, it will assist and help people make better decisions faster with richer insights that shouldn’t be a distraction from their main purpose: discovering the right people faster and reducing time-to-hire.
Whether you’re trying to fill a niche role or meet aggressive hiring targets, Findem’s Copilot for Sourcing will help you find the precise candidates you need. The platform uses attributes, a new type of talent data generated from 100,000+ public sources, that allow you to look up candidate profiles like you’re running a Google search.
Instead of relying on keywords or Boolean strings, Findem's 1M+ attributes enable you to find profiles of people who are “relationship builders,” “one of the first 50 hires,” or had a “fast career progression.”
Request a demo if you would like to see how Findem makes this possible.
Start with the warmest leads
.png)
