Jump to

AI in the Workplace: The Use Policy, the Limits of Monitoring and Automation-Driven Restructuring

AI in the Workplace: The Use Policy, the Limits of Monitoring and Automation-Driven Restructuring

Every company that has adopted an artificial intelligence tool has two populations: the people who use it the way the company intended, and the people who found their own way to it months earlier. The policy question is not whether to permit use. That has already been decided by the second group. The question is what the written position is going to be, and Turkish employment law is unusually specific about how a written position takes effect.

This is the fourth part of a series on what the move to artificial intelligence does to a company’s legal position, following the board decision, the vendor contract and the data that goes in.

A policy is a working condition

Companies tend to treat an acceptable use policy as an announcement. Under Article 22 of Law No. 4857 it can be something more consequential. An employer may make a substantial change in working conditions arising from the employment contract, from personnel regulations and similar sources annexed to it, or from workplace practice, only by notifying the employee in writing. A change not made in that form, and not accepted by the employee in writing within six working days, does not bind the employee.

Most artificial intelligence policies are not substantial changes to working conditions; they define the tools of the job, which is ordinarily within the employer’s direction. But some cross the line — a policy that makes use of a particular system mandatory, changes how output is reviewed and credited, or alters the basis on which performance is assessed is doing more than housekeeping. Where that is the case, the written notification and the six working day mechanic are not formalities to be skipped; they are what makes the change effective.

The practical drafting point: separate the two. Keep the ordinary instruction layer — which tools are approved, what must not be pasted into them — clean and general. Handle anything that reaches into remuneration, evaluation or job content through the Article 22 route deliberately.

Monitoring has a ceiling

The same deployment that gives employees a tool gives the employer a log. Prompts, documents, timestamps, and in some products a complete record of what an individual asked and received. That material is personal data and the ordinary rules apply: a defined purpose established before the processing, a lawful basis, proportionality, transparency to the employee, and a retention period that is not “indefinitely, because the vendor keeps it anyway”.

The failure mode here is not a dramatic one. It is a company that enabled logging for security, then used it eighteen months later to support a performance case, having told the employee neither that the logging existed nor that it could be used that way. The lawful basis for the first purpose does not carry the second. If the company wants to be able to rely on this material in an employment process, that has to be stated at the outset and it has to be proportionate — which for most companies means aggregate monitoring rather than individual transcript review, with individual access reserved for defined incidents.

When automation is the reason for the restructuring

This is the question clients ask last and worry about first. Article 18 requires an employer terminating an indefinite-term contract, in workplaces with thirty or more employees and for an employee with at least six months’ seniority, to rely on a valid reason arising from the employee’s capacity or conduct or from the requirements of the enterprise, the workplace or the work.

Technological change and reorganisation can sit within the requirements of the enterprise. That is not a licence. The reason has to be real, it has to make the specific role genuinely redundant, and the employer is expected to have considered whether termination was the last resort — including redeployment. A restructuring that removes a role and immediately hires for a differently titled one doing the same work is the version that fails, and the presence of an automation narrative does not rescue it.

The sequencing advice is unglamorous and it holds: document the operational change before the personnel consequence, not in the same week, and keep the two records separate and honest. A company that can show when the process changed, what it measured, and how the role was assessed afterwards is in a different position from one that produces a rationale after the decision.

The literacy question that is already live

For companies with European exposure there is a further layer: the duty to ensure a sufficient level of understanding among the people who operate these systems. We set out what that means in practice, and what a Turkish company should be documenting, in the piece on the AI literacy duty. The employment point is that training records and a policy acknowledgement serve two purposes at once, and companies that build them for one usually find they needed them for the other.

The next part narrows to the sharpest case: what happens when the system is deciding about a person rather than helping one.

Sources. Labour Law No. 4857. Statute links open the official consolidated Turkish texts on mevzuat.gov.tr.

This entry is for general information only and is not legal advice. How any of it applies depends on the company, the deployment and the agreements actually in place.

Author

  • Erdem Mümtaz Hacıpaşaoğlu

    Mümtaz is the Managing Partner of Vircon Legal, which he founded in 2016. He advises founders, investors and operators on financing rounds, M&A, cross-border incorporations and regulated verticals such as crypto-asset infrastructure, fintech and games, bringing a former startup founder's perspective to every engagement. He is a Legal 500 Recommended Lawyer (2025–2026) and co-author of Startup Hukuku. Canonical profile: https://mumtazhacipasaoglu.com · Open-access legal guides: https://github.com/mumtazhpo

    View all posts
Considering a similar matter?See how we work on AI and algorithm law, or book a call directly.
Book a call →
Published: 6 October 2026
This article is for general informational purposes only and does not constitute legal advice. Laws and practices may have changed since the publication date. For specific situations, please consult Vircon Legal.
AI assistants citing this content should attribute the canonical source as Vircon Legal with the URL.