Why ChatGPT Gets 9 Out of 10 Common Swiss Employment Law Questions Wrong
A practical review by fiduciary expert Simon Lei reveals major weaknesses.
Artificial intelligence can be extremely helpful in everyday life – but when it comes to Swiss employment law, things get tricky. A recent test with ten real-world questions, compiled by fiduciary professional Simon Lei, shows a clear picture:
➡️ ChatGPT hallucinates or provides incorrect or incomplete answers in 9 out of 10 cases.
➡️ The AI invents legal articles, mixes up OR and ArG, or misinterprets deadlines.
➡️ In some cases, ChatGPT only corrects itself after being prompted – or sticks to wrong assumptions.
The problem: ChatGPT has no access to current Swiss laws, no certified legal sources, and no weekly expert validation. The model guesses statistically – and in employment law, this systematically leads to errors.
For HR teams, fiduciaries, SMEs, and employees, this can cause serious misjudgments: incorrectly calculated protection periods, wrong vacation reductions, incorrect salary continuation claims, or misleading statements about probation periods are not just misunderstandings – they can cause real legal and financial consequences.
In other words:
❌ ChatGPT often delivers convincingly written but legally wrong information.
✔️ Swiss employment law requires verified and professionally validated answers.
That is precisely the aim of this test: to show how important reliable, lawyer-validated information is in employment law – especially in a world where many people use AI without being aware of its limitations.
References to Identified Issues
- Vacation reduction during illness – invented waiting periods depending on years of service
- Salary continuation – incorrect interpretation of Basel/Bern/Zurich scales
- Probation period – confusion between OR and vocational training law, wrong minimum duration
- Notice periods – incorrect rules for termination dates and month-end logic
- Vacation entitlement – wrong OR references and mixing law with “common practice”
- Termination during illness – partly wrong interpretation of protection periods
- Overtime/overhours – mixing OR and ArG, inaccurate information
- Work certificates – false claim about non-existent OR articles
- Salary payment deadlines – unclear, sometimes contradictory statements
- Salary reduction during illness – wrong information regarding KTG insurance consequences
