If you've applied for a job in the last two years, there's a good chance an algorithm looked at your application before a person did. AI screening now handles an estimated 60–80% of the early hiring funnel at large employers — yet most candidates, and many HR professionals, couldn't precisely explain what it does.
This guide breaks down what AI screening actually is, the different forms it takes, and what it means for both sides of the hiring table.
AI Screening, Defined
AI screening is the use of automated systems — often machine learning models — to evaluate, rank, or filter job candidates before a human recruiter reviews them. It can happen at multiple stages: resume parsing, skills assessments, video interview scoring, and chatbot-based pre-qualification.
The common thread is that a candidate's progress to the next stage depends, at least in part, on a model's output rather than a person's judgment.
The Main Types of AI Screening
Resume and application screening — software parses resumes for keywords, experience matches, and qualifications, ranking or filtering candidates before a recruiter sees them.
Asynchronous video interview scoring — platforms like HireVue analyse recorded responses to set questions and produce a score or ranking.
Conversational pre-screening — chat-based assistants (like Paradox's Olivia) ask qualifying questions in real time and route candidates accordingly.
Skills and game-based assessments — automated tests that score technical or cognitive ability, often used as an early-stage filter.
Why Employers Use It
The case for AI screening is mostly about scale. A role that attracts thousands of applicants can't be manually reviewed in any reasonable timeframe — automated screening is, in practice, the only way to make that volume tractable. Done well, it can also reduce the variance that comes from different recruiters applying different standards.
Why It's Controversial
The risks fall into two buckets. The first is bias — models trained on historical hiring data can reproduce and scale whatever bias existed in that data, as we covered in our analysis of the Eightfold AI and Workday lawsuits.
The second is opacity. Candidates are often screened out without knowing AI was involved, why they were rejected, or how to appeal — issues we explored in our look at why candidates are walking out of AI interviews.
AI screening isn't inherently good or bad — it's a force multiplier for whatever process it's added to.
What Candidates Should Know
If you're applying for roles at large companies, assume some form of AI screening is involved at the resume stage at minimum. Tailoring your resume to match the language of the job description (without keyword-stuffing) genuinely helps with parsing-based systems. For video assessments, treating them with the same preparation as a live interview — clear audio, good lighting, structured answers — matters because the scoring model is evaluating clarity and content, not "vibes."
What Employers Should Know
Three practices separate AI screening done well from AI screening that creates legal and reputational risk: disclosing AI use to candidates upfront, keeping a human in the loop before final rejection, and periodically auditing the system for disparate impact across protected groups — now a legal requirement in some jurisdictions.
The Bottom Line
AI screening is no longer an emerging trend — it's existing infrastructure that most candidates have already encountered, often without realising it. Understanding how it works is now a practical necessity for job seekers, and getting the implementation right is increasingly a legal and competitive necessity for employers.
Sharingan AI publishes independent research on how AI hiring tools actually work — not just how they're marketed — so both sides of the hiring process can navigate it with clearer eyes.