Introduction
AI continues to transform software testing, offering new opportunities to improve efficiency and coverage. As its capabilities expand, it’s essential to use AI responsibly - supporting human testers, protecting data, and complying with evolving regulations.
AI Potential in Manual Functional Testing
AI has proven to be a game-changer in the field of software testing, offering a new approach to manual functional testing. Good prompt can create a user flow path that can help newbies channel their testing to some specific areas based on the scope.
Let's see an example of the prompt and the response from AI.
Hello, ChatGPT! Can you help me create the user flow for manual functional testing of the e-commerce website that has the following features: Homepage, Account, Wishlist, Product Overview Page, Product Detail Page, Cart, Checkout, Newsletter, Footer, and Header? Also, make sure that placing orders is out of scope. Testers are not allowed to place any orders while they are allowed to proceed to the Checkout. Bug types that are in the scope of the test are Functional (Low, High, and Critical), Content, and Visual.
With a prompt like this, you don't give AI any data about the customer or any other specifics from the test cycle.
Now, let's see the response from the AI.
Sure. In that case, the user flow would look like this:
1. Homepage:
- Check if all the graphics, images, and texts load properly on the Homepage.
- Ensure that clickables (buttons, dropdowns, sliders) are functional.
2. Header and Footer:
- Test that all the links in the header & footer are directed to the correct page.
- Check if the company logo redirects to the home page.
- Verify the contact information and social media links working properly.
- Check if all the graphics, images, and texts load properly.
3. User Account:
- Test the functionality of creating a new account.
- Confirm if the form accepts/rejects valid/invalid email addresses.
- Check the login function with the correct/incorrect username and password.
- Test for recovering the forgotten password.
- Verify the editing functionality in account details - change password, change email, and delivery address.
- Check if all the graphics, images, and texts load properly.
4. Newsletter:
- Confirm the subscription and unsubscription functionality.
- Confirm if the form accepts/rejects valid/invalid email addresses.
- Check if all the graphics, images, and texts load properly.
5. Product Overview Page:
- Ensure that all products are displayed accurately with correct descriptions and prices.
- Check the function of the filter and sort on the page.
- Verify the 'Add to Cart' and 'Add to Wishlist' buttons functioning properly.
- Check if all the graphics, images, and texts load properly.
6. Product Detail Page:
- Verify that each product detail page correctly portrays the product image, price, description, specifications, and customer reviews if available.
- Ensure the 'Add to cart' and 'Add to wishlist' functionality is working as expected.
- Check if all the graphics, images, and texts load properly.
7. Wishlist:
- Check the functionality to add/remove items from the Wishlist.
- Verify the functionality to move Wishlist items to the Shopping Cart.
- Check if all the graphics, images, and texts load properly.
8. Shopping Cart:
- Verify the function of adding, updating, and removing products from the cart.
- Check updates in the total price with changes in the products in the cart.
- Check if all the graphics, images, and texts load properly.
9. Checkout Process:
- Test entering delivery addresses and selecting delivery options.
- Proceed to the payment gateway process but do not complete the order.
- Check if all the graphics, images, and texts load properly.
10. General:
- Check on various devices & browsers (cross-browser test and mobile responsive check).
Keep in mind that testers are not allowed to place any orders, hence actual transactions should not be performed.
Important: Based on the test instructions, feature descriptions, and OOS section, the prompt must be edited for the best possible results of the output. If you still wonder if the strategy given by the AI complies with the test cycle scope, reach out to the test cycle TL for help.
However, Test IO testers need to keep in mind that AI should not replace human judgment. The combination of AI's data-driven insights with human testers' intuition and creativity is crucial to a comprehensive testing process.
Responsible Usage Principles
1. Data Privacy
Always use anonymized or synthetic data when possible.
Avoid sharing sensitive information with AI tools.
Ensure compliance with privacy regulations (e.g., GDPR).
2. Fairness and Bias
Train AI models on diverse datasets to reduce bias.
Regularly review AI-generated test documentation for fairness.
3. Transparency
Document how AI is used in your testing process.
Make AI-driven decisions explainable to stakeholders.
4. Human Oversight
Keep humans in the loop for critical decisions.
Establish clear escalation paths if AI outputs are unclear or potentially harmful.
Regulatory Considerations
Stay informed about new regulations such as the EU AI Act and NIST AI Risk Management Framework. These require documentation, risk assessment, and human oversight for high-risk AI systems.
Best Practices Checklist
Use diverse, representative data for AI training.
Regularly audit AI models for bias and fairness.
Document AI decision-making processes.
Ensure human oversight and clear escalation paths.
Protect sensitive data and comply with privacy laws.
Stay informed about regulatory changes.
Continuously monitor AI systems in production.
Conclusion
Responsible usage of AI in testing is a dynamic, multi-faceted challenge. By embracing fairness, transparency, accountability, and compliance, organizations can harness AI’s full potential while safeguarding quality, ethics, and trust in their software products.