AI-Generated Seed Data and Test Fixtures

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ai-coding-claude-database-seed-data-creators

1. When using AI to generate seed data for a database with foreign key relationships, which capability represents a key advantage?

   1. Converting the seed data into compiled machine code
   2. Ensuring every generated email matches a real person's inbox
   3. Generating rows that maintain referential consistency across joined tables
   4. Automatically deploying the database to production

2. A developer asks AI to generate 100 user records with obviously fictional names. Which naming pattern would best meet this requirement?

   1. Sequential numbered IDs like 'User001', 'User002'
   2. Actual historical figures like 'George Washington'
   3. color-space-noun format like 'Azure Telescope'
   4. Random GUIDs like 'a3f9-8b22-0014'

3. What risk remains even when using AI to generate test data?

   1. The AI will always generate syntactically invalid SQL
   2. The AI cannot generate any numeric values
   3. The AI will inevitably format dates incorrectly
   4. The AI might inadvertently include memorized real names or email addresses

4. Property-based generators for fuzz testing are best described as:

   1. Manual datasets created by hand for regression testing
   2. Code that automatically produces many randomized inputs to test edge cases
   3. Database backup procedures that preserve state
   4. Pre-written test scripts that run in sequence

5. If a developer wants AI-generated test data that mirrors their production environment's characteristics, what should they provide to the AI?

   1. A detailed story about their company culture
   2. A copy of their entire production database
   3. A list of all real customer names and addresses
   4. Sample data showing the distribution and shape of production values

6. Which of the following represents an edge case that AI-generated seed data should explicitly include?

   1. Data that exactly matches the tester's personal information
   2. Empty strings, unicode characters, and extremely large numbers
   3. Records with completely random binary blobs
   4. Only typical values that appear 99% of the time

7. A database has tables for 'authors' and 'books' with a foreign key relationship. When generating seed data, what must be true about these tables?

   1. Every book record must reference an existing author ID
   2. Foreign keys should be omitted from test data
   3. Books must always be generated before authors
   4. Authors and books must use identical primary keys

8. The primary purpose of test fixtures (seed data) in software development is to:

   1. Generate revenue through data marketplace sales
   2. Provide consistent, known data for reproducible test execution
   3. Replace the need for writing any test cases
   4. Ensure the production database always has fresh data

9. Which scenario best illustrates the privacy concern with AI-generated seed data?

   1. An AI uses outdated programming language syntax
   2. An AI model outputs a real person's actual email address in generated test data
   3. An AI fails to generate enough test records
   4. An AI generates a database schema incorrectly

10. Fuzz testing primarily aims to discover:

    1. Performance improvements in database query speed
    2. Unexpected failures when inputs include malformed, random, or extreme values
    3. Security vulnerabilities in network firewalls
    4. Memory leaks in graphical user interfaces

11. Why might using real customer data as seed data be problematic even if the data is 'anonymized'?

    1. AI might have memorized the original data and could regenerate it
    2. Real data cannot be used in testing environments
    3. Anonymization removes all useful characteristics
    4. Anonymized data always fails database constraints

12. A developer wants to test how their application handles usernames containing emojis and non-Latin characters. What should the seed data include?

    1. Strings that are exactly 8 characters long
    2. Unicode characters like '用户123' or 'José😀'
    3. Only ASCII letters and numbers
    4. Usernames that match famous social media accounts

13. When AI generates seed data without any sample data provided, what limitation typically occurs?

    1. The AI will refuse to generate any data
    2. The database will become corrupted
    3. The generated data may not match the real production distribution
    4. The generated data will always be 100% accurate

14. Test fixtures differ from production data in that fixtures should be:

    1. Updated daily to reflect current business trends
    2. Generated only once and never modified
    3. Consistent across test runs and free of real personal information
    4. Loaded with the most recent production updates

15. A junior developer generates seed data and notices many records contain recognizable company names from their industry. What should they conclude?

    1. The seed data is now production-ready
    2. The database foreign keys are working correctly
    3. The AI likely memorized these from training data and this poses a privacy risk
    4. They should increase the number of records to 10,000