Can you trust AI for advice on your rights at work?
ChatGPT and other AI tools can help you understand general legal questions about your job. But when it comes to your specific rights or a workplace dispute, it’s important to know their limitations.
Imagine this situation: You have been dismissed and do not understand why. You want to know what rules apply to dismissals and what rights you have.
You need help, so you ask ChatGPT, Claude or another AI chatbot.
You quickly get an explanation of a few legal concepts and some suggestions for questions that may be relevant to explore.
You ask the chatbot to help you write an email to your manager in which you make certain demands regarding your dismissal. Naturally, it is happy to help.
But this is where things go wrong.
The chatbot writes a polished email, and everything sounds convincing. The problem is that the content of the email does not actually make sense – and it is almost impossible for you to spot this because you are not an expert in the legal terminology the chatbot uses.
Your manager quickly dismisses the email and rejects your demands. You are told that there is nothing more to negotiate, and you are now in an even worse position than before you asked the AI chatbot for help.
AI can’t do everything
At Pharmadanmark’s Legal Department, we often speak to members who have used AI to look into their rights. We also find that the answers do not always provide a sufficient basis for assessing the case.
AI can do a lot, but it can’t do everything. Here, we will try to help you understand when you can use AI – and when you should be cautious.
And remember: We always recommend contacting one of our legal advisers if you have questions about your employment, if you are dismissed, or if you are involved in a conflict with your workplace.
Get help from Pharmadanmark
A simple rule of thumb: Never rely on an AI-generated answer alone when making decisions in a specific case.
Remember that our legal advisers are always ready to help assess your particular situation. Get in touch with us so we can help you decide what to do next.
Phone: +45 39 46 36 05
Mail: jura@pharmadanmark.dk
Use AI for general questions
An answer from an AI chatbot is not the same as a legal assessment of your specific situation.
However, you can use AI to learn more about general questions. For example, if you have received a warning, you can use AI to explore questions such as:
What is a written warning?
What consequences can a warning have?
What should I be aware of?
What questions might be relevant to ask my employer?
You can also ask about terms in your employment contract or general rules concerning, for example, notice periods, holidays or restrictive covenants.
But: Be aware that the answer may be inaccurate, lack important context or be based on outdated information.
Your specific case depends on the details
Things become more complicated if you ask: “Was my dismissal unfair?” or “Am I entitled to compensation?”
The answer depends on many different details:
What does your contract say?
Are you covered by a collective agreement?
What happened in the period leading up to the dismissal?
What emails have been exchanged?
Have you previously received a warning?
You have probably not described every relevant circumstance in your question to the AI chatbot, which means it cannot give you a fully accurate and precise answer.
AI can ask follow-up questions, but you cannot be sure that the chatbot will identify every factor that matters in your particular case.
It is also important to remember that a detail that may seem insignificant can change the entire legal assessment.
Our legal advisers know which details matter and how to identify gaps in a case. That is why it is important to contact us if you are dealing with a specific issue that requires legal assistance.
AI only knows the version you give it
An AI chatbot always bases its response on the information you provide.
For example, if you describe a conflict at your workplace, the chatbot will probably receive an account influenced by the fact that you are in a difficult situation, as emotions often run high during a conflict.
That is completely natural. But when making a legal assessment, it is also important to examine the situation from other perspectives:
What would the employer say?
What documentation is available?
Are there emails or other information that point in a different direction?
There can also be a difference between experiencing something as unfair and having a legal claim.
An important part of our legal advice is therefore to ask questions that challenge the case. Not because we are taking the employer’s side, but because we want the assessment to be based on the strongest possible foundation.
Be careful what you share
An employment-related case may contain personal data, confidential company information or information about colleagues and managers.
You should therefore think carefully about what information and documents you share with an AI chatbot.
Once you have shared information, you cannot simply take it back – and it can be extremely difficult to have information deleted from AI services.
Remember that you do not need to upload your entire case to a chatbot to get a general overview. Instead, keep your questions as general as possible if you want to use AI to understand a legal concept or set of rules.
Check legal sources
An AI-generated answer may include references to legislation, court rulings and other legal sources. This does not necessarily mean that the information is correct.
AI may misrepresent a source or, in some cases, refer to a court ruling or another source that does not exist.
It can be difficult to tell whether the information is accurate or fabricated. You should therefore always ask us for advice to make sure that the information you have found is correct.
Do you need legal assistance?
You can always contact our legal team if you have questions or need help.
Phone: +45 39 46 36 05
Mail: jura@pharmadanmark.dk