An AI interview is a job interview in which software asks the questions, scores the answers, or does both. It is usually the first round interview: the candidate answers on video or in chat with no recruiter present, and the system rates each answer against the job's criteria. Recruiters then see the scores and decide who moves on.
The main formats
Employers use the term for three main formats.
| Format | Who asks | What the candidate does | What gets rated | A vendor's own description |
|---|---|---|---|---|
| Recorded one-way video | Questions shown on screen, set in advance | Records an answer to each question, at a time they choose | The recorded answers, rated by software, by reviewers, or both | HireVue: "One-way asynchronous interviewing giving candidates 24/7 flexibility" |
| AI-led conversational interview | An AI interviewer that speaks and asks follow-ups | Holds a two-way conversation | Each response, against criteria for the role | HireVue: AI agents ask "probing follow-up questions, in a natural two-way conversation" |
| Text or chat screen | A chatbot | Types answers to qualification questions | Whether stated requirements are met | Paradox: "Validate important qualifications upfront, through chat or text" |
A live video call with a human interviewer, where software only transcribes or summarises, is a different thing. HireVue's video interviewing page, which covers recorded and live interviews, lists "AI-powered interview summaries, searchable transcripts". In a live call a person still runs the conversation. The one-way format has its own guide: one-way video interviews.
What is an AI interview used for in hiring?
It sits at the screening stage, before you meet anyone from the team. HireVue's candidate FAQ says its interviews and assessments "are intended to be first steps in the overall interviewing process" and are "basically just replacing the typical 20-minute screening phone call with a recruiter". The same FAQ says the top group then moves "on to person-to-person interviews before hiring decisions are made by people."
That is the vendor's account of how its customers use it. An employer can set up its process differently, so ask what happens after the interview. What usually comes next is covered in what happens after an interview.
What an AI interview scores
The vendors quoted here describe scoring what the candidate says, rated against a rubric for the role. HireVue's AI Interviewer page says the AI "evaluates each response against your role-specific criteria, highlights whether it meets them, and explains why", and lists a "Custom 1-5 rubric" among its scoring options.
HireVue once analysed video as well. In a press release dated 12 January 2021, HireVue said that "early in 2020, HireVue proactively removed the visual analysis component from all of its new assessments". It gave the reason that advances in natural language processing "had significantly increased the predictive power of language", so "visual analysis no longer significantly added value to assessments."
HireVue's candidate FAQ also says:
- "Hirevue never uses the video recording of your interview in order to identify you, which is sometimes referred to as 'facial recognition.'"
- "Assessments never look at age, race, gender, or even things on your resume such as GPA or the college you attended."
- "Only skills and competencies are measured."
Those are one vendor's statements about its own product, and other vendors and employers can make different choices. Before an AI interview, it is reasonable to ask the employer four things: what is scored, what is not, whether a person reviews the result, and whether you can ask for an accommodation. HireVue's FAQ says candidates "can also request accommodations from your recruiter or the hiring team".
Is an AI interview legal? What regulators require
The rules ask for different things. Illinois and Singapore's PDPC focus on telling candidates what happens to their data and getting consent. New York City requires bias audits. Hong Kong's PCPD wants a person in control of high-risk decisions, and the EU treats hiring AI as high-risk and bans emotion recognition at work.
| Place | Rule | What it says |
|---|---|---|
| Illinois, US | Artificial Intelligence Video Interview Act, 820 ILCS 42 (in force 1 January 2020) | Before the interview, the employer must tell the applicant AI may be used, explain "how the artificial intelligence works and what general types of characteristics it uses", and get consent. Videos are deleted within 30 days of the applicant's request. |
| New York City, US | Local Law 144 of 2021 (enforced from 5 July 2023) | An automated employment decision tool needs a bias audit "within one year of the use of the tool", a public summary of the audit, and notices to candidates. DCWP says the notice must come "10 business days prior to use". |
| European Union | AI Act, Regulation (EU) 2024/1689 | "AI tools for employment" are high-risk, for example "CV-sorting software for recruitment". Obligations for these Annex III uses start on 2 December 2027. Bans "emotion recognition in workplaces and education institutions". |
| Hong Kong | PCPD, Artificial Intelligence: Model Personal Data Protection Framework (June 2024) | Lists "Assessment of job applicants" among AI uses that may be higher risk, and says a high-risk system should take a "human-in-the-loop" approach. |
| Singapore | PDPC, Advisory Guidelines on use of Personal Data in AI Recommendation and Decision Systems (issued 1 March 2024) | Under the PDPA, consent is generally needed to use personal data in AI systems, and people must be told the purpose when consent is sought. |
Illinois and New York City
The Illinois Act, as published by the Illinois General Assembly, also limits who sees the video: "persons whose expertise or technology is necessary in order to evaluate an applicant's fitness for a position." Since 1 January 2022, an employer that "relies solely upon an artificial intelligence analysis" to choose who gets an in-person interview must report applicants' race and ethnicity to the state each year, so the state can report "whether the data discloses a racial bias".
In New York City, a candidate can report an employer that skipped the bias audit, the audit summary or the notices to DCWP, and the page says "You do NOT need to create an account to file a complaint".
European Union
The European Commission's AI Act page (last updated 3 August 2026) says the high-risk dates for Annex III uses were moved to 2 December 2027 by the "AI Omnibus" amendment. The prohibited practices have applied since 2 February 2025.
Hong Kong and Singapore
Both documents are guidance from the data protection regulators on how their laws apply to AI. The PCPD framework says "human actors should be held accountable for the decisions and output made by AI." The PDPC guidelines say notifications "should be proportionate to the risks of each use-case" and encourage organisations to explain how personal data is used and which features are more likely to influence the result.
This is general information, not legal advice; check the official source for your case.
How ROUND.1 runs its AI interview
ROUND.1 is a hiring marketplace for Hong Kong and Singapore. Each candidate takes one AI first-round video interview, in a web browser with a camera and a microphone; there is no audio-only or phone interview. It lasts about 20 minutes by default (roles can be shorter), 30 at most, and each role is interviewed in one language, English or Mandarin Chinese. Before it starts, the candidate consents to the processing of their video and voice under Hong Kong's PDPO and Singapore's PDPA.
The interview is scored competency by competency, and each score is backed by quotes from the transcript. Accent, pronunciation, grammar, fluency, personality traits and protected characteristics (such as age, gender, race, religion, disability or marital status) are not scored. There is one attempt, and the interview cannot be restarted. If the connection drops, the same interview can resume, once by default; a role can allow more resumes or none. A resume continues the interview and does not start it again.
A company first sees the scored evaluation without the candidate's name or contact details, which open when it shortlists the candidate. A person at the company decides whom to hire. The full list of facts and limits is on ROUND.1, explained, and the practical side is in how to prepare for an AI interview.
Frequently asked questions
Is an AI interview fair?
It depends on the tool and on how the employer uses it. Some rules exist to check: New York City's Local Law 144 requires a bias audit within one year of using an automated employment decision tool, and Illinois requires race and ethnicity reporting where an employer relies solely on AI to pick who gets an in-person interview. HireVue says its assessments never look at age, race or gender. Ask the employer what is scored and whether a person reviews the result.
Is an AI interview legal?
In general yes, with conditions in some places. Illinois requires notice, an explanation and consent before an AI-analysed video interview. New York City requires a bias audit and notices for automated hiring tools. The EU AI Act treats AI used in recruitment as high-risk. Hong Kong's PCPD and Singapore's PDPC have published AI guidance under their personal data laws. This is general information, not legal advice; check the official source for your case.
Can I retake an AI interview?
It depends on the employer and the vendor, so ask before you start. At ROUND.1 there is one attempt and the interview cannot be restarted. If the connection drops, the same interview can resume, once by default, and a role can allow more resumes or none. A resume continues the interview where it stopped and does not start it again.
What is a first round interview?
A first round interview is the first conversation after you apply. It checks the basics of the role before longer interviews with the hiring team. Traditionally it is a short screening call with a recruiter. HireVue's candidate FAQ describes its interviews as replacing the typical 20-minute screening phone call with a recruiter, before the top candidates move on to person-to-person interviews. An AI interview can fill this slot.
Does a human see my AI interview?
It depends on the employer. HireVue's AI Interviewer page says recruiters always make the final call. Illinois law limits sharing of AI-analysed interview videos to people whose expertise or technology is needed to evaluate the applicant. At ROUND.1 a company first sees a scored evaluation without your name or contact details, which open when it shortlists you, and a person at the company decides whom to hire.
Sources (10)
- HireVue — Frequently Asked Questions for Candidatesaccessed 2026-10-14
- HireVue — AI Intervieweraccessed 2026-10-14
- HireVue — Video Interviewingaccessed 2026-10-14
- Paradox — home page (applicant screening)accessed 2026-10-14
