Many users wonder how accurate Jurilo really is – especially compared to well-known AI systems like ChatGPT or Claude.
We wanted to find out exactly and therefore conducted a comprehensive practical test.
The Test
We used 60 real questions from Swiss employment law – the same questions we also use internally for ongoing quality control.
The answers were reviewed by the renowned law firm Streiff von Kaenel, which specializes in employment law and validates our content weekly.
The evaluation was based on a 10-point system, taking into account both factual accuracy and legal precision.
Average score:
Claude: 6.95 / 10
ChatGPT: 7.20 / 10
Jurilo: 9.80 / 10
Verified by the employment law firm Streiff von Kaenel.
The Result
Jurilo is clearly at the top with an accuracy of 98%.
While general language models like ChatGPT or Claude can provide useful background information, they lack the legal context and Switzerland-specific legal basis (CO, Labor Act, collective labor agreements, court rulings, etc.).
Jurilo, on the other hand:
is based on verified answers from Swiss lawyers,
is reviewed and updated weekly,
understands legal contexts and exceptions,
provides precise source references (article numbers and court cases).
Conclusion
Artificial intelligence can assist with legal questions – but law requires precision.
With its specialization in Swiss employment law, Jurilo achieves a quality that even the most advanced language models currently cannot offer.
➡️ Try Jurilo now for free
Experience the difference between generative AI and a platform specifically trained for Swiss law.
Note on the Image
An interesting phenomenon emerged when creating the comparison graphic:
ChatGPT, which was used to visualize the bar chart, displayed the value 7.2 as smaller than 6.9 despite multiple correction instructions.
An illustrative example that even advanced language models occasionally "hallucinate" or misinterpret simple logical relationships.
Further proof of why Jurilo relies on verified, tested, and legally reviewed answers – instead of generic AI responses.
