Remote hiring has transformed access to technical talent. It has also created a new category of hiring risk. Over the past year, fake and AI-assisted applications have moved from an occasional problem to a recurring feature of technology recruitment, and the most organised schemes are now the subject of international security advisories.

This article sets out what’s driving the trend, the risks it creates for the business and the controls technology leaders should expect their hiring process to include.

Why are fake job applicants becoming more common?

Three trends have come together. Remote hiring has removed many of the face-to-face checks that used to catch impostors. Generative AI makes a convincing CV, cover letter or interview answer quick and cheap to produce. And organised fraud, including state-backed schemes, now uses fake identities to win technology jobs and access company systems.

Volume is part of the problem. Reporting on the ACCA’s latest global talent trends survey, Personnel Today notes that some candidates are using AI chatbots to apply, adding to application volumes. In the same survey, 69% of UK finance and accountancy leaders said they had concerns about AI being used in hiring. When every application is well written, the usual signals that separate strong candidates from weak ones become less reliable.

At the extreme end, hiring has become a route for state-backed fraud. In September, agencies in Australia, Germany, Japan and the US issued a joint advisory on a North Korean campaign that infected more than 30,000 devices through bogus coding tests. As The Register reported, this comes on top of North Korea’s better-known tactic of placing its own IT workers in technology roles under fake identities. Researchers estimate that roughly 100,000 are employed or seeking work worldwide.

For leadership teams, the implication is that recruitment now forms part of the organisation’s security perimeter.

How common are fake job applicants?

More common than most hiring managers assume, and hard to measure, because the cases that succeed go unnoticed. Early data from a fraud-detection tool Checkr is trialling flagged 36% of applicants for review, rising to half for software and AI engineering roles. A flag isn’t proof of fraud, but it shows how many applications need a second look.

The practical consequence is that screening takes longer and carries more weight. Application volumes are rising at the same time as written applications have become a weaker guide to ability.

What do fake applications look like in practice?

Most follow a few patterns. A CV written by AI that describes experience the candidate can’t discuss. Interview answers fed in real time by an AI tool. Face-swapping software on video calls. And, once hired, someone other than the successful candidate doing the work, sometimes through laptop farms that disguise where they’re based.

Not every case is fraud in the criminal sense. Some candidates exaggerate, and AI has made that easier. Others use proxies, with a stronger engineer taking the technical interview and someone else starting the job. At the serious end, the person hired is a front for someone else entirely.

In the cases The ONE Group’s Technology team has encountered, the discrepancy became clear only when the candidate was asked to attend in person and discuss their work in detail.

What is the risk to the business?

The cost goes well beyond a bad hire. Someone in a technical role may have access to code, customer data and systems from their first week. And employers who hire someone without the right to work, without correct checks, face civil penalties of up to £45,000 per illegal worker, or £60,000 for repeat breaches.

There are also the familiar costs of a failed appointment: lost months, a disrupted team and a search that has to restart. In a market where experienced engineers are already scarce, those delays affect delivery directly.

Is it a problem if candidates use AI?

No, not on its own. Plenty of good candidates use AI to tidy up a CV or prepare for an interview. It becomes a problem when someone uses it to pretend to be a person they’re not, with experience they don’t have or answers they can’t give without help.

Developers take a pragmatic view. In Stack Overflow’s 2026 developer survey of more than 30,000 people, nearly half said they trust AI when they can verify its work. That’s a sound principle for hiring too.

We’d recommend setting a clear policy on acceptable AI use at each stage, including take-home tests and live exercises, and sharing it with candidates in advance. Clear rules make breaches easier to identify and are fairer to legitimate applicants.

What are the warning signs of a fake applicant?

Look for a pattern. Common warning signs include CV claims that don’t hold up in interview, repeated refusals to meet in person, interruptions or background voices on video calls, visual glitches that may point to face-swapping software, and requests to be paid in cryptocurrency.

Each of these can have an innocent explanation. Together, they warrant closer scrutiny before an offer is made.

What controls should a remote hiring process include?

Verification at every stage. Before interview, check that the CV, online presence and first conversation tell the same story. During interview, keep cameras on and probe detail. Before an offer, verify identity and right to work and take references directly. After they start, confirm the person logging in is the person you hired.

In practice, the most effective controls are straightforward:

• Depth over breadth in interview. Explore one project in detail, including what went wrong and who else was involved. Detail is difficult to fabricate for long.
• Test reasoning. Ask why decisions were made. Scripted or AI-assisted answers tend to break down away from the obvious response.
• Screen-shared walkthroughs. Ask candidates to explain work they’ve built. Live explanation of their own work is far harder to fake.
• At least one in-person stage. For hardware, embedded and test roles, this is often part of the job anyway.
• Independent references. Use contact details sourced by your business, not supplied by the candidate.
• Post-offer checks. Send equipment to a verified address and confirm identity at onboarding.

What should technology leaders do now?

Treat hiring integrity as a shared responsibility between technology, HR and security. Review where identity is verified in your process, who owns each check, how AI use is governed, and what happens if concerns arise after someone has started. Most businesses can close the main gaps without slowing the process for strong candidates.

Verification tools are improving and have a place, combining document checks, biometrics and online-presence signals. They work best as one layer in a process that still puts an experienced person in front of every candidate.

How can The ONE Group help?

The ONE Group’s Technology team recruits software, data, infrastructure and cyber security specialists across the UK, the US and Europe. Every candidate is interviewed by one of our consultants and has their technical knowledge tested before they’re introduced to a client, and wherever possible we meet them in person. We also advise on interview structure and on verification checks proportionate to each role.

To discuss your technology hiring or review your remote hiring process, contact The ONE Group on 01733 234000 or hello@theonegroup.co.uk.

FAQs

Can AI detection tools spot fake candidates?

They can help. Verification tools now combine document checks, biometrics and online-presence signals, and they’re useful for managing volume. But a flag isn’t proof, and tools can get it wrong in both directions. They work best alongside a structured interview with detailed follow-up questions, not instead of one.

Should we ban candidates from using AI?

Probably not. Many strong candidates use AI to refine a CV or prepare for interview, and many developers use it daily at work. It’s more effective to define what’s acceptable at each stage, such as take-home tests or live exercises, and tell candidates before they start.

Which roles are most at risk of fake applicants?

Remote technical roles, particularly software and AI engineering, where early verification data shows the highest flag rates. Roles with access to sensitive systems or data are attractive targets, and state-backed schemes have focused on IT positions. Hardware and test roles tend to be lower risk because they usually involve on-site work.

What should we do if we suspect a candidate isn’t who they say they are?

Pause before making an offer. Request an in-person or on-camera meeting, verify identity and right to work through official checks, and take up references using contact details you’ve sourced yourself. If the person has already started, involve your IT security team promptly, because access to your systems is the main risk.