Last month, Anthropic, an AI company, put out a message to potential job applicants, urging them to refrain from using AI bots in the application process.
A message on its online application form states: “While we encourage people to use AI systems during their role to help them work faster and more effectively, please do not use AI assistants during the application process. We want to understand your personal interest in Anthropic without mediation through an AI system and we also want to evaluate your non-AI-assisted communication skills.”
While an AI company banning job prospects from using GenAI might seem ironic, it’s an understandable response to an increasingly common problem that’s creating endless headaches for hiring managers.
A tidal wave of AI applications
Two-thirds (65%) of jobseekers are using AI tools to assist with their applications, according to Career Group, a recruitment company. AI is commonly used to write cover letters, for instance, enabling candidates to send out many more applications than ever before – and in far less time.
Natalie Burns, strategy partner at UnitedUs, a branding agency, says her company was recently “inundated” with applications for a senior strategist role. The number of applicants for the role was 173% higher than for a similar job advert posted the previous year.
While Burns acknowledges the change is partly down to market conditions – job vacancies in the UK have steadily declined since 2022, meaning there are more people competing for fewer positions – she also believes access to AI assistants is partly to blame.
“A lot of it stems from the use of automated technologies and AI, which make it quicker for applicants to churn out vast quantities of applications,” Burns says.
AI makes it quicker for applicants to churn out vast quantities of applications
Another survey, by the software company Capterra, found that 41% of UK jobseekers had used AI to lie about their skills in a job application, while 27% admitted to using AI to complete a test or skills assessment.
For Burns, findings such as these are damaging her trust in candidate submissions. “It’s difficult to discern the contribution and influence of the person you’re hiring,” she adds.
Michael Kahn, chief of staff at the London-based PR firm Rostrum, says jobseekers’ reliance on AI is also leading to lower-quality cover letters. Many applications now sound “eerily similar” to one another, he claims, and it isn’t difficult to identify the ones written by AI.
“They lack the personal touch of a real applicant,” Khan says. “They may be beautifully written and very polished, but they sound robotic.”
Caroline Fischer, UK country manager at Welcome to the Jungle, a careers site, adds that some candidates even neglect to delete the prompt response from the AI chatbot. ChatGPT, for instance, precedes its response to being asked to write a job application with, “Sure! Here’s a polished and compelling job application”.
“It’s a reflection of how somebody might do their day-to-day job – without critique or creativity,” she says. “In a competitive and noisy job market, they are unlikely to be the candidate that progresses to interview.”
Filtering out the noise
Although some employers are forbidding AI assistance in the application process, ensuring candidates follow this rule is a challenge. “I don’t think people will stop using AI tools, because Pandora’s box has been opened,” says Khyati Sundaram, CEO of Applied, a recruitment platform.
AI-detection tools can help recruiters filter out some AI-written applications. However, they are not yet capable of identifying robot-written text with complete accuracy.
People won’t stop using AI tools, Pandora’s box has been opened
Sundaram suggests companies rethink their recruitment processes. “There are artifacts of the recruitment industry that we’ve struggled to move away from, such as the CV and cover letter,” she says.
One solution is to switch to skills-based hiring. By focusing on an individual’s skills, rather than the experience and qualifications on their CV, companies can assess candidates more accurately and objectively, Sundaram says. “About 80% of hiring is the human element and I don’t think AI can replace that,” she adds.
But not all skills assessments are immune to AI infiltration either. Rostrum’s assessment process, for instance, used to include a written test. This task however has been “corrupted” by candidates routinely using AI programmes to complete it, Kahn explains.
He now requires candidates to complete the written portion of the application either in Rostrum’s office or over a video call.
The AI gender gap
The failure to filter out AI-generated applications could also impact the workplace gender balance. A 2024 study identified an AI gender gap, with 50% of men having used AI in the past 12 months compared with 37% of women. This means that, thanks to GenAI, employers could be receiving more applications from men than from women.
Virgin Media O2 was aware of this potential problem when it redesigned its apprenticeship application process this year to be more “AI resilient”. Its head of future careers, Karen Handley, explains: “Whether you use AI or not, we want everyone to be treated equally in the process, so we try to make sure we don’t favour people that use AI and remove barriers for those who may not have access to this technology.”
The application process for its apprenticeship roles focus on potential, strength and motivation. Handly says: “We’ve changed some of the exercises in our assessment centre so they focus more on future strengths, as opposed to competency-based questions, which are easier to ask AI systems to answer.”
Video interviews have also been replaced with telephone interviews, as there are fewer ways to utilise AI over the phone, she adds.
With the use of AI in the application process becoming more prevalent, candidates who chose to write their own cover letters may now hold an advantage. “It’s become rare to receive an application that has not been touched by AI, so writing your own cover letter and CV is now the differentiator,” Khan says. “It really does help people stand out.”
For employers, using a variety of assessment methods, which can test for soft skills as well as technical skills, is the best way to filter out AI-supported applicants. “The more information you have about candidates, the more nuanced you can be about the job match,” Sundaram says.
AI may have made it easier to apply for jobs but bringing the human element back into the recruitment process can help employers get a better idea of who they’re really hiring.
Last month, Anthropic, an AI company, put out a message to potential job applicants, urging them to refrain from using AI bots in the application process.
A message on its online application form states: “While we encourage people to use AI systems during their role to help them work faster and more effectively, please do not use AI assistants during the application process. We want to understand your personal interest in Anthropic without mediation through an AI system and we also want to evaluate your non-AI-assisted communication skills.”
While an AI company banning job prospects from using GenAI might seem ironic, it’s an understandable response to an increasingly common problem that’s creating endless headaches for hiring managers.
