How it works
Interviews run on Capifiq, the meeting platform built for them. Because the interview and the integrity check are the same session, Capifiq verifies the environment in real time and delivers timestamped, tamper-evident proof of what happened: no bolted-on layer, no behavioral guesswork, no lockdown on the candidate.
The problem
A subscription industry now sells AI assistance directly to candidates, built for one purpose: to be invisible during a live interview. These tools render answers on a transparent overlay only the candidate can see, listen to the interviewer through the machine's own audio, or run a second assistant in a browser tab, and generate an expert-sounding response in a second or two. Many are marketed, explicitly, as "undetectable" on Zoom, Teams, and Meet.
It is not a fringe problem, and it is not slowing down. The economics are one-sided: a monthly subscription costs less than dinner, and the prize is a six-figure role. The category keeps growing, and every new tool that ships widens the surface a hiring team has to defend.
The blind spot
The defenses most teams rely on were designed for a different threat, and the newest tools are built specifically to slip past all three:
Screen sharing only transmits the surface the operating system is told to share. A tool that draws its overlay outside that surface, or runs on a second device entirely, never enters the stream your interviewer sees.
Browser proctoring lives inside the browser tab. It can watch tab switches and copy-paste, but it has no window into native desktop applications or the wider system the interview runs on, and today's overlays are native desktop applications, invisible to anything confined to a browser tab.
Behavioral analysis infers cheating from gaze, posture, and response timing. It is indirect by nature: it can flag an honest, nervous candidate and still miss a careful one, and it adapts more slowly than the tools it is chasing.
What the usual defenses miss
Capifiq takes a different position. Instead of inferring intent from behavior or matching a list of known product names, it verifies the interview environment directly and records what it finds as evidence, coverage that does not depend on the specific tool already being known.
Schedule the interview in Capifiq, invite the candidate, monitor integrity live, review any findings, and finish with a sealed report.
Add the role, the panel, and the time. Capifiq schedules the interview and creates the protected session it runs in. There is no external meeting link to paste.
Capifiq emails the candidate clear instructions and a secure link to join the interview in the Capifiq client. The integrity check comes with it. There is nothing separate to install.
While the interview is running, the interviewer can confirm session integrity and watch verified events appear on the live dashboard.
When Capifiq surfaces an event, the interviewer sees it in the moment and decides how to proceed with the conversation. Dispositioning the record afterwards is a separate authorised reviewer's job.
At the end of the interview, Capifiq creates an integrity report with the session timeline, findings, timestamps, and sealed evidence.
a3f6b7c9e2d4…Coverage
Capifiq is designed to surface integrity-relevant activity running on the device. These are the common vectors, not a fixed list. Each describes what a candidate can abuse during an interview.
Desktop and web tools that transcribe the interview and surface generated answers on an overlay only the candidate sees.
AnyDesk, TeamViewer, Chrome Remote Desktop and similar tools that let another person view or drive the candidate's machine.
VMware, VirtualBox and emulators used to isolate the visible interview from a second workspace running unauthorized tools.
OBS, ManyCam and injected feeds used to replace the live webcam with pre-recorded or manipulated video.
Extensions and web apps that listen to the interview and generate suggested answers in a side panel or background tab.
Selective capture, overlays, extra displays and virtual-display tricks that present a clean shared screen while hiding content elsewhere.
Coverage is not a list of product names. Because it verifies the environment rather than matching known binaries, Capifiq surfaces this behavior even when the specific tool is renamed, privately built, or brand new, the kind a signature-based scanner has not catalogued yet.
The interview record
Nothing to invite into the call, nothing to switch on. It just happens, every time.
The interview is captured as it happens, so nothing important comes down to memory or hurried notes. The shared pads are part of that recording, so a candidate's work doesn't disappear when the call ends.
The conversation becomes a searchable, timestamped transcript you can revisit and share with the panel. Captions run live so nobody has to strain to follow, and a caption stays a caption, with nothing read into it.
The rhythm of question and answer across the interview is part of the record, not just the words.
Recording, transcript and the session timeline share one clock, so anything worth revisiting can be found at the point in the conversation it belongs to.
It points. A person decides. Capifiq does not grade answers, rank candidates, or produce a score. Everything here is an observation tied to the moment in the conversation it refers to, and candidates have documented rights and a review process.
Write code together, or sketch the system on a shared whiteboard. No screen sharing, and no scratch file that vanishes when the call ends. Both are captured in the recording.
Nothing written on a pad is graded or scored, and we never read anything into how fast someone types. Candidates who think differently, or type differently, don't come out looking suspicious for it.
The evidence
Behavioral tools hand you a suspicion score and ask you to trust it. That puts the hardest judgment, was this person actually cheating?: back on the interviewer, with nothing solid to stand on.
Capifiq is built the opposite way. A finding is deterministic: the condition was present in the session, or it was not. Each one is timestamped, hashed at capture, and sealed into the report, so it holds up when a hiring decision is questioned. You are not asked to interpret a number. You are handed a record.
Every finding carries
Privacy by design
Capifiq documents exactly what can be captured during a monitored interview, which actions the interviewer controls, and how long records are kept. Collection happens only during the interview session: nothing before it starts or after it ends.
Every interview
Interviews run in Capifiq and are recorded, including the session screens, for the length of that session.
Interviewer controlled
The interviewer can capture a screenshot of any display at any moment, whether it is currently being shared or not.
Always captured
Control-key combinations and monitored hotkey activity are captured throughout the active interview.
Enterprise retention
Interview data, including sealed integrity evidence, is retained to match your enterprise retention policy, for review and dispute resolution.
We observe and report. We never decide. It does not block, terminate, or restrict the candidate's machine, and it runs only inside the scheduled interview window. Findings go to the interviewer; a person makes the call.
Questions
Get started
Run a scoped pilot on your own interviews and judge the evidence for yourself. No credit card.
Free white paper
Fifteen pages on how live AI assistance works, why conventional proctoring misses it, and what actually protects your interviews: written for hiring leaders.