Last updated: January 2025
Fastly AI uses artificial intelligence to analyze vehicle photos and provide automotive diagnostics and maintenance recommendations. Please read this disclaimer to understand the capabilities and limitations of our AI technology.
Nature of AI Recommendations
Fastly AI' diagnostics, maintenance predictions, and recommendations are generated by machine learning algorithms trained on automotive data. While we strive for accuracy, our AI:
- Provides probabilistic assessments, not definitive diagnoses
- May not identify all possible conditions or factors
- Works best with clear, well-lit vehicle photos
- May be less accurate for rare or unusual vehicle issues
- Cannot physically examine vehicles or perform physical inspections
Not a Replacement for Professional Advice
Fastly AI is designed to augment—not replace—professional automotive expertise. Our recommendations should not be considered a substitute for:
- Professional certified automotive technicians and ASE-certified mechanics
- Factory-trained specialists
- Dealership service departments
- Professional diagnostic equipment testing
For high-value vehicles, complex mechanical issues, or safety-critical repairs, we recommend consulting with qualified automotive professionals in addition to using Fastly AI.
Repair Recommendations
Our repair and maintenance recommendations are based on common automotive repair practices and available research. However:
- Effectiveness may vary based on local conditions, vehicle make/model, and driving patterns
- Some repairs or parts may vary in availability by region
- Always read and follow manufacturer specifications and repair procedures
- Consider safety implications and local inspection requirements
- Get a second opinion for major repairs when possible
Accuracy and Continuous Improvement
Our AI models are continuously trained and improved. Current accuracy rates are:
- Common mechanical issues: ~96% accuracy
- Warning light diagnosis: ~94% accuracy
- Wear assessment: ~92% accuracy
- Repair cost estimates: within ~20% of actual (varies by region and shop)
- Rare or complex issues: Lower accuracy rates
These figures are based on internal testing and may vary in real-world field conditions.
Photo Quality Impact
Diagnostic accuracy depends heavily on photo quality. For best results:
- Use good natural lighting
- Focus clearly on the problem area
- Include close-up and wider context shots of the issue
- Avoid blurry or dark images
- Show both damaged and normal components for comparison
Limitation of Liability
Fastly AI and its creators are not liable for:
- Vehicle damage or repair costs resulting from following our recommendations
- Incorrect diagnoses or missed issues
- Adverse effects from suggested repairs or maintenance
- Economic losses related to vehicle damage or repair costs
Users assume full responsibility for decisions made based on Fastly AI' recommendations.
Feedback and Reporting
If you believe our AI provided an incorrect diagnosis or recommendation, please report it through the app. Your feedback helps us improve accuracy for everyone.
Contact
Questions about our AI technology? Contact us at:
brian@fastlyai.org