AI has had something of a seismic effect on job applications: in some cases I would go as far as saying it has now destroyed an assumption that a CV can be relied on to filter and assess candidates for a role.
Take a recent post our sister company recruited for. We are working on a project to source a machine learning engineer (ironically!) for a government project we are invested in. We received over 200 applications for the post, and following the first wave of recruitment it was clear that a) we could not rely on the answers given on application forms and b) we could not rely on the content of CVs.
We interviewed candidates who quite clearly had asked Chat GPT to transform their CV to give them the best chance of a job interview. Some even clearly used ChatGPT during the interview to answer questions (the first round of interviews was via Zoom). The majority of candidates had very limited technical abilities. Their experience was not evident when they completed a practical test on a face to face basis. It made the whole process a complete waste of everyone’s time and the project now requires a second round of recruitment.
Employers and HR professionals are all beginning to get used to seeing AI CVs, which will usually be stuffed with keywords and buzz words Chat GPT has determined need to be included.
In larger organisations for some time now algorithms manage, filter, and determine which candidates are shortlisted. A lot of recruiters in sectors involving the numbers game (eg IT recruitment where there can be 10,000 applications for 1 role) rely on these systems via keyword matching and pattern recognition. Unfortunately AI now seems to have rendered some of this filtering a bit redundant, as candidates are using AI to “game” the system by optimising their CVs & applications to get past the algorithms. Manual CV reviewing may become more important again in these sectors we think.
The traditional CV, with its chronological career structure and standardised layout, has existed relatively unchanged for decades. The simple format gives the reader the ability to quickly skim read through and decide on whether to read on or dismiss. To a certain extent, law is a bit sheltered from damage by AI to its recruitment processes. If a candidate needs to be a 5 year PQE employment law solicitor in order to apply for a role, they cannot use AI to engineer their way in to a job interview. However, at a more junior level (for example paralegal) it can make quite a significant difference to the decision making process and has the potential to waste an awful lot of time.
Practical Alternatives
We think that manual CV reviews for all job applications are going to be essential, in order to filter out the AI CVs and avoid wasting time. Similarly, a face to face technical test at the outset for some roles may be essential. Other options include skills-based assessments – asking a candidate to look at a file and comment on the contents and what they would do next is a good and quick example of a way to assess ability. Digital portfolios have been talked about – suggestions that candidates produce a video talking about themselves (ie a CV but by the person). Simulation – you could ask members of staff to be clients and conduct a mock meeting to see how the candidate performs in a simulated environment.
Conclusion
The transformation sparked by AI’s disruptive entrance into recruitment is irreversible and it will be interesting to see whether firms start to use videos, simulations or skill assessments at the outset instead of bothering with written applications & CVs. At present for senior roles it probably has a very limited effect, but for support staff and junior posts we foresee problems ahead for HR & recruiters!
PS if you are planning on using AI to write a job application or CV, please don’t. We can spot them from a mile away and they read very poorly!
