AI risks and safe use
Understand the limitations and risks of AI-generated code and content, and how to use bldrAgent responsibly.
AI-generated content: what you should know
bldrAgent uses AI to generate code, data models, and content. Like all AI systems, it has limitations. Understanding these helps you get better results and avoid potential issues.
Known limitations
AI can make mistakes
The AI may:
- Generate code that looks correct but has subtle bugs
- Misunderstand ambiguous instructions
- Produce different results for the same prompt on different attempts
- Miss edge cases in complex logic
Mitigation: Always test your app thoroughly before publishing. Use the preview mode to check behaviour across user flows.
AI output is not always optimal
The AI optimises for correctness and completeness, not necessarily for performance or elegance. Very large, complex apps may benefit from review by an experienced developer.
Mitigation: If you need expert review, hire a bldrDeveloper.
AI may produce outdated information
The AI's knowledge has a training cut-off date. For integrations with third-party services, API documentation may have changed since the AI was trained.
Mitigation: Always verify integration configuration against the current documentation of the third-party service.
Content safety
bldrAgent's AI is configured to:
- Refuse requests to generate malicious code, phishing pages, or content that violates our Terms of Service
- Decline to produce content that is harmful, abusive, or illegal
- Not exfiltrate private data from your workspace
If you encounter AI-generated content that seems harmful or inappropriate, report it to safety@bldr.app.
Accuracy of AI-generated data
If you ask the AI to generate sample or seed data, it creates plausible-looking but fictional content. Do not use AI-generated data as a replacement for real research or verified information in production apps.
Your responsibilities as the app creator
When you publish an app built with bldrAgent, you are responsible for:
- Ensuring the app complies with applicable laws (GDPR, CCPA, accessibility standards, etc.)
- Reviewing and testing AI-generated code before publishing
- The accuracy of content generated for your app
- Obtaining proper licences for any third-party content used
bldrAgent provides the tools; you are responsible for what you build with them.
Reporting concerns
If you have concerns about AI-generated content or potential misuse of the platform:
- Email safety@bldr.app
- Use the Report button in the AI chat (three dots → Report)