How we use AI.
The recruitment industry is in the middle of an AI inflection. Some firms are using it to scale practices we think should never be scaled. This page is what AEY & Associates does and does not use AI for, in plain language, kept up to date and reviewed every six months.
Last reviewed · 14 May 2026
As a faster typist.
-
Research and market mapping
Identifying candidate universes, cross-referencing public sources, building structured talent maps from large unstructured datasets. AI accelerates the legwork; the judgement remains ours.
-
Internal drafting and operations
Drafting structured documents, summarising long meeting notes, generating first-pass written outputs that we then edit, fact-check and rewrite before they leave the firm. Treated as a faster typist, not as a thinker.
-
Sector research and benchmarking
Compiling background research on companies, sectors and technologies; cross-checking compensation benchmarks against publicly available data. Always verified by hand against AEY placement records before any figure is shared with a client.
-
Coordination and scheduling
Standard productivity tooling that uses AI under the hood - calendar suggestions, email triage, transcription of meetings (only with consent). These are treated as the same category as any other tool.
Never as a substitute for judgement.
-
Candidate assessment
Every assessment note in every shortlist is written by a human at AEY who has spoken with the candidate. AI is not used to generate, rank, score or filter candidate assessments. The judgement we are paid for is the judgement of the named people delivering the search - never automated.
-
Shortlist composition
The decision to include or exclude any candidate from a shortlist is made by a human. Tools that "auto-shortlist" by matching CVs to JDs are not used. Pattern-matching is the failure mode we are paid to avoid, not to scale.
-
Candidate outreach
Every initial approach to a candidate is written by a human. We do not use AI-generated outreach. The senior people we approach receive a high volume of automated recruiter messages; ours are deliberately not part of that volume.
-
Confidential client information
No client name, brief detail, candidate name, salary information or other confidential information is ever shared with public AI tools. Where we use AI internally, it is via tooling that has been vetted for data handling - and never with information covered by a confidentiality agreement.
Three commitments that govern the above.
- 01
Disclosure on request
If a client or candidate asks whether AI was used in any part of an engagement, we will answer honestly and specifically. We will name the tool, the task and the human who reviewed the output.
- 02
No surreptitious automation
You will never receive an email from "Farid" that was not written or edited by Farid. The firm's name is on the door; the writing on it is ours.
- 03
Reviewed every six months
AI tooling is moving faster than any other category of professional-services tool we have seen. We review and update this page every six months - or sooner if a material change in our practice warrants it.
AI is going to reshape recruitment more than it has reshaped any professional-services category we have lived through. The firms that win will not be the ones that automate the work - they will be the ones that automate the operations and re-invest the saved time in the judgement that clients are paying for.
That is the model we are running, and the model this page commits to. If you have a question about how it applies to a specific engagement, ask. We will answer specifically.
If something on this page isn't clear, ask.
Transparency only works if it can be tested. Write directly to Farid and we will respond - usually within one business day - with the specific answer for your situation.