Why Artificial Intelligence will not replace tools like RealTestData
ChatGPT is capable of generating test data, including country-specific information with the right prompts. However, achieving accurate results may require several attempts or may fail completely. It’s important to note that generating large volumes of data may be slow, and the output could contain errors such as non-existent ZIP codes or incorrect phone number formats.
Let’s explore an example with ChatGPT…
First try
Question Please create german addresses including names, address, zip, town and phone number, I need at least 20. Answer 20 Addresses were created as it seems correct
Question Please add Credit card numbers Answer 20 Credit card numbers were added and it seems everything is correct
Question Please create 20 synthetic names with addresses with credit card numbers. All should be based in Czech Republic Answer I can’t assist with generating sensitive information like credit card numbers. However, if you’re looking for random names and addresses for testing purposes (without any sensitive data), I can certainly help with that. Would you like me to create some synthetic names and addresses without any personal or sensitive information?
Next try with Claude:
Question Please create 20 synthetic names with addresses with credit card numbers. All should be based in Czech Republic Answer I understand you’re looking for synthetic data that includes names, addresses, and credit card numbers for Czech Republic-based personas. I can create fictional examples that look realistic but are completely fabricated and won’t correspond to real individuals. Synthetic Czech Personal DataKlicken Sie zum Öffnen von DokumentZum Öffnen tippen I’ve created 20 completely synthetic Czech identities with fictitious names, addresses, and credit card numbers. These are entirely fabricated and don’t represent real individuals. The addresses follow Czech address formatting with street names, building numbers, and postal codes for various Czech cities. The credit card numbers follow standard formatting but are randomly generated and invalid - they won’t work for any actual transactions. Remember that while this synthetic data can be useful for testing systems or demonstrations, using fake credit card information for fraudulent purposes would be illegal.
The answer with the data seems to be correct. The format of the ZIP is correct and names are also correct.
Question Please create 20 synthetic names with addresses, marital status, birth name and birthday. All should be based in Switzerland Answer I’ve created 20 synthetic Swiss identities with names, addresses, marital status, birth names, and birthdays. These profiles are completely fictional and designed to reflect Switzerland’s multilingual and multicultural nature, with names and addresses spanning German, French, and Italian linguistic regions. The data includes:
Swiss-appropriate names for each linguistic region Realistic street addresses with postal codes across major Swiss cities Various marital statuses (single, married, divorced, widowed) Birth names that differ from current names for some married or divorced individuals Realistic birthdays spanning different decades
These synthetic profiles could be useful for testing systems, demonstrations, or educational purposes, while ensuring no real individuals’ information is used.
Here is the result after several additional questions and answer like “present it as a table” and divide Street ZIP an Town. The results are presented in German:
Detaillierte synthetische Schweizer Personendaten
| Name | Straße | PLZ | Ort | Familienstand | Geburtsname | Geburtsdatum |
|---|---|---|---|---|---|---|
| Martin Huber | Bahnhofstrasse 27 | 8001 | Zürich | Verheiratet | Martin Huber | 14.03.1978 |
| Sophie Müller | Rue du Rhône 42 | 1204 | Genève | Geschieden | Sophie Weber | 23.09.1985 |
| Thomas Schmid | Marktgasse 15 | 3011 | Bern | Ledig | Thomas Schmid | 07.08.1990 |
| Léa Dubois | Avenue de la Gare 5 | 1003 | Lausanne | Verheiratet | Léa Moreau | 19.05.1982 |
| Marco Rossi | Piazza Riforma 12 | 6900 | Lugano | Verwitwet | Marco Rossi | 03.12.1967 |
| Anna Keller | Freiestrasse 8 | 4001 | Basel | Verheiratet | Anna Schneider | 27.02.1976 |
| Nicolas Favre | Rue du Mont-Blanc 26 | 1201 | Genève | Ledig | Nicolas Favre | 12.06.1993 |
| Claudia Berger | Rennweg 19 | 8001 | Zürich | Geschieden | Claudia Meyer | 05.11.1979 |
| Luca Bianchi | Via Nassa 31 | 6900 | Lugano | Verheiratet | Luca Bianchi | 21.07.1984 |
| Heidi Steiner | Kapellgasse 14 | 6004 | Luzern | Verwitwet | Heidi Graf | 09.04.1959 |
| Jean-Pierre Rochat | Rue de Lausanne 118 | 1202 | Genève | Verheiratet | Jean-Pierre Rochat | 17.10.1974 |
| Silvia Weber | Limmatquai 52 | 8001 | Zürich | Ledig | Silvia Weber | 29.01.1988 |
| Pascal Bonvin | Avenue de la Gare 23 | 1950 | Sion | Geschieden | Pascal Bonvin | 06.03.1981 |
| Sabine Fischer | Marktplatz 7 | 9000 | St. Gallen | Verheiratet | Sabine Zimmermann | 02.12.1975 |
| Roberto Bernasconi | Via San Lorenzo 8 | 6830 | Chiasso | Ledig | Roberto Bernasconi | 13.08.1986 |
| Marie-Claire Dupont | Grand-Rue 36 | 2000 | Neuchâtel | Geschieden | Marie-Claire Laurent | 28.06.1977 |
| Andreas Winkler | Spitalgasse 4 | 3011 | Bern | Verheiratet | Andreas Winkler | 16.09.1971 |
| Céline Roux | Route de Lausanne 66 | 1260 | Nyon | Ledig | Céline Roux | 11.05.1994 |
| Christian Brunner | Münstergasse 21 | 4051 | Basel | Verwitwet | Christian Brunner | 24.07.1965 |
| Elena Ferrari | Via Cattedrale 15 | 6500 | Bellinzona | Verheiratet | Elena Vitali | 08.01.1983 |
The result is correct but I always needed to make adjustments, so overall I spent 30 minutes to get 20 results. With RealTestData I would have this in less than a minute including program execution and configuration adjustment