6 in 10 Overstate Their AI Skills: How to Assess Real AI Competency

According to a survey by hiring verification platform GCheck, 63% of full-time US workers have exaggerated or misrepresented their AI skills, and among Gen Z the figure reaches 80%.

Of those who listed AI skills, only 34% said they could actually handle work at a professional level. An “AI skills bubble,” in which simple prompting experience gets dressed up as problem-solving ability, is spreading across the job market.

Now that generative AI is within everyone’s reach, companies have started treating AI skills as a core hiring criterion. Yet most lack the standards and methods to objectively verify what candidates can really do.

What companies need now is a verification system that runs from resume screening all the way through the interview.

3 Ways to Cut Through Inflated AI Claims and Find Real Competency

Separating real competency from impressive-sounding AI experience on paper is harder than it looks. Without clear evaluation criteria, a candidate’s actual AI literacy and ability to solve problems and verify output end up being judged on an individual interviewer’s gut feel. That’s why you need verification guidelines that HR and hiring managers can agree on, backed by a step-by-step system that covers screening, interviews, and data management.

Step 1. Resume Screening: Check the Depth of Skills and the Role Context

Resume screening starts with setting verification criteria that fit the type of hiring process, whether that’s blind hiring, large-scale open recruitment, or year-round and ad hoc hiring.

The key is having criteria that tell you whether the AI experience on an application is inflated casual use or real, job-ready command of the tools. Look past statements that a candidate has used a tool and check their practical understanding across the whole workflow, from defining the problem to using data and verifying results. That lets you filter for candidates who meet your organization’s standards, even in a large pool of applications.

✅ Resume and Personal Statement Review in Telta Interview Pro
Telta, an AI-powered business and HR insights service built on the mental models* of HR experts, analyzes even large volumes of applications quickly and precisely. It goes beyond keyword search to give you a full picture of each candidate’s practical skills, role context, and AI literacy, so you can review candidates against your organization’s standards quickly and accurately.

- What is a mental model? It is a structured system that captures how HR experts evaluate candidates and make intuitive judgments. Telta has turned the evaluation know-how developed by HRD and organizational assessment experts from Korea’s leading conglomerates into algorithms, so you can review applications with AI more precisely and quickly.

Step 2. Interview Evaluation: Ask Questions That Test Verification Skills

At the interview stage, we recommend finding out whether a candidate simply accepts whatever AI produces or can critically verify and control it.

You can design questions in three directions: how candidates choose tools and apply them to their work (practical use), how they spot and handle errors in AI output (critical verification), and how they protect internal data and respect copyright (risk and ethics).

✅ Telta’s Tailored Interview Questions for Each Candidate
Based on its analysis of each application, Telta suggests tailored interview questions that precisely test a candidate’s AI skills and ability to handle errors. This helps take interviewer subjectivity out of the process and verify candidates’ critical thinking and risk management skills.

Step 3. Data-Driven Verification: Quantify Competencies and Turn Them into Organizational Assets

Rather than relying on interviewers’ fragmented memories or subjective notes, interview evaluations should be standardized into objective data through real-time conversation analysis and a competency assessment framework.

A real-time interview support setup, with automatic transcription, summaries of key answers, and detection of positive and negative competency signals, lets interviewers focus fully on what the candidate is saying and on testing their competencies. Going further, a data-driven analysis process that scores role-specific and common competencies and grounds each rating in what the candidate actually said reduces rating gaps between interviewers. It also lets you manage candidates’ real competencies as objective metrics and as part of your organization’s data assets.

✅ Interview Pro, Telta’s AI Interview Support Solution
During the interview, it transcribes and summarizes what is said in real time and detects key competency signals, so interviewers can focus entirely on verification. After the interview, it generates a quantified competency report from the conversation data, reducing rating gaps and helping you manage all of your hiring data as an organizational asset.

‍

Fast, Accurate Screening at Scale with Telta

Accurately identifying candidates’ AI skills matters, but it is just as important to build a system that verifies them consistently across the organization. This is especially true in high-volume settings such as blind hiring or open recruitment at large enterprises, where it is hard to fairly verify, in limited time, which of countless applicants meet your organization’s standards.

Telta, an AI-powered business and HR insights service, analyzes the flood of applications from open recruitment to ad hoc hiring with AI built on the mental models of HR experts, supporting precise resume screening and tailored interview questions for each candidate. And during interviews, its Interview Pro feature supports interviewers in real time, removing subjective judgment and recording the entire interview as objective, data-driven evaluation metrics your organization can build on.

Start identifying candidates’ real competencies, with evidence, from the screening stage on with Telta.

‍

‍