AI-enhanced Persona-based Test Case creation derived from Product Requirements Document

Mikołaj Kostyrko
19 min readFeb 10, 2024

· Introduction
· Person-based testing
· The app — case study
· Prompting — persona creation step
· PRD — Meal Plan Sharing
· Requirements conversion into Test Cases
· Traceability Matrix creation populated with personas
· Functionality (PRD) enhancement
· Summary

Introduction

Creating a Product Requirements Document (PRD) is a crucial step in any IT project. This document serves as the blueprint for the entire development process of a given feature or the project phase. It typically begins with a comprehensive analysis of business needs and user requirements. Once these are defined, the PRD outlines the project scope, features, functionalities, and technical specifications in a clear and organized manner. By providing a detailed roadmap, the PRD ensures that all stakeholders have a common understanding of the project’s objectives and helps guide the development team throughout the project lifecycle.

The role of a Quality Assurance engineer in the creation of a Product Requirements Document is crucial for ensuring that the document accurately captures all necessary quality criteria and considerations. This boils down to focusing on:

  • analyze the requirements to ensure they are clear, complete, and testable. Any ambiguities or inconsistencies are flagged for clarification.
  • identification of Non-Functional Requirements - In addition to functional requirements i.e. reliability, scalability, security, and maintainability.
  • risk assessment - identification of potential risks that may impact the quality of the product. They document these risks in the PRD and propose mitigation strategies to address them.
  • accessibility and usability considerations - ensure that the PRD includes considerations for accessibility and usability. They verify that the product will be usable by individuals with disabilities and that it meets relevant usability standards.
  • validation and verification criteria — to define validation and verification criteria for each requirement outlined in the PRD. Validation criteria specify how the product will be validated against user needs, while verification criteria outline how each requirement will be verified during testing.
  • traceability matrix — a creation of a traceability matrix that links each requirement in the PRD to corresponding test cases. This ensures that all requirements are adequately tested and helps track the test coverage throughout the development process.

In this article, we will focus on the last of the mentioned aspects: the traceability matrix or rather test cases that take a different perspective based on created personas.

Person-based testing

What is persona-based testing? Persona-based testing is a testing approach that focuses on the behavior, preferences, and needs of different user personas when validating software products. A persona represents a fictional character created to embody a specific user group’s characteristics, goals, and challenges. Persona-based testing involves designing test scenarios, cases, and strategies based on these personas to ensure that the software meets the diverse needs of its users.

Here’s how persona-based testing typically works:

> Persona Creation [QA teams collaborate with stakeholders, including product managers, designers, and marketers, to create personas representing different user segments]

> Understanding User Behavior [ Each persona is defined with specific attributes, such as demographics, goals, motivations, challenges, and preferred workflows. This understanding helps QA teams empathize with users and anticipate how they would interact with the software]

> Test Scenario Design [Test scenarios are designed and cases tailored to each persona’s characteristics and use cases. These scenarios simulate real-world interactions with the software, covering various user journeys, tasks, and workflows.]

> Test Data Generation [Test data is generated or selected to reflect the characteristics and preferences of each persona. This ensures that the test environment closely mirrors the actual usage scenarios and provides realistic testing conditions.]

> Test Execution [Test cases are executed against the software, with each scenario representing a different persona’s perspective. QA engineers verify that the software behaves as expected for each persona, addressing any deviations or issues discovered during testing.]

The app — case study

The presented concept should apply to any app (or at least most of them), here I will choose a Carb Manager App > https://www.carbmanager.com/ app [Keto & Low-Carb Made Easy / The #1 Low-Carb and Keto Diet App] — for no particular reason, but health and diet oriented app tend to gather different crowd and thus I believe the idea presented here will be easier to exemplify.

Prompting — persona creation step

Once we covered the basics we can focus on the persons and test case creation with the aid of an AI i.e. ChatGPT or a similar LLM which is based on prompting interaction.

Responses presented below were created using ChatGPT (3.5).

On a daily basis, I use the so-called RODES Prompt Framework (structured upon R — Role, O — Objective, D — Details, E — Examples, S — Sense Check) which in my opinion is the most effective single/zero prompting technique (for more info check: https://www.thepromptwarrior.com/p/5-prompt-frameworks-level-prompts).

The role is known: QA engineer

Objective: persona creation

Examples: are taken from a blog post by Katrina the Tester on testing persona generation: http://katrinatester.blogspot.com/2015/01/generic-testing-personas.html

Following this perspective, the prompt should look as follows:

You are a senior software QA specializing in test case creation and catching inconsistencies in product requirement documents.

Personas are a tool to adopt the habits and feelings of different users. During testing, they can be used to discover different types of problems.

Create five testing personas that would use Carb Manager App (Keto & Low-Carb Made Easy / The #1 Low-Carb and Keto Diet App)

Here are four generic testing personas that you could adopt when completing a testing task:

Manager Maria
Maria is a busy executive who interacts with the application between meetings. She is impatient and often not focused on her task, completing activities in haste. Maria will:
Stick to the quickest workflow through the application
Use shortcut keys
Fill in the minimum number of fields to get a result
Make mistakes in her efforts to get things completed quickly
Require fast responses and may repeat an action if the application takes too long to respond
Often be called to a meeting midway through a task

Hipster Hilary
Hilary likes to investigate new functionality and areas of the application that are outside of the mainstream. She is an early adopter and an avid explorer. Hilary will:
Investigate new features as soon as they become available
Explore all possible paths through a workflow to determine which she prefers
Frequently use areas of the application that are less popular
Have unusual data input compared to other clients e.g. different units of measure
Be accessing the application from an unusual browser, operating system or device

Careful Claire
Claire enjoys routine. She uses the same workflows each time that she interacts with the application, taking care to ensure that she is consistent and the information she provides is complete. Claire will:
Stick to popular features of the application
Notice and investigate any visible changes to these features, e.g. a new button is added
Complete every field possible when entering information
Be verbose when asked to enter notes of her own, e.g. a reason for editing a record.
Be patient with long response times

Sneaky Shirley
Shirley likes to break things. She knows about common security holes in software and likes to explores the applications that she uses to feel confident about their ability to protect her information. Shirley will:
Enter SQL and JavaScript injection in to application input fields
Manipulate URLs to attempt to access private information
Violate constraints on input fields by entering invalid information
Try to generate as many error messages as possible

International Ioana
Ioana is on an overseas vacation. She periodically uses the application for specific reasons, e.g. to retrieve a piece of information or complete a single task. Ioana will:
Use the application outside of local business hours
Be accessing the application from multiple locations and time zones
Use a variety of browsers, operating systems and devices
Occasionally have poor network connectivity that is slow and unreliable
Be using a variety of keyboard layouts
Enter personal information that includes foreign language characters

Elder Elisabeth
Elisabeth is of an older generation with relatively little knowledge of computing. She has trouble understanding many software applications. Elisabeth will:
Use the application slowly, taking time to read each screen
Frequently use the 'Back' button to remind her previous information
Have the interface font of the application enlarged via browser settings or zoom
Require simple and clear interfaces in order to successfully complete a task
Seek out online help to assist her
Be using outdated technology including an older browser and operating system

Here is the response I was provided with (GPT- 3.5)

Here are five personas tailored to testing the Carb Manager App:

Busy Bethany:

Description: Bethany is a full-time working mother of two who uses the Carb Manager App to manage her family's meal plans and track their nutrition intake. She's always on the go, balancing work, household chores, and family responsibilities.
Testing Approach:
Focuses on efficiency and ease of use.
Prefers straightforward workflows with minimal steps.
Values quick access to essential features like meal logging and recipe management.
Requires intuitive navigation to find information quickly amidst her busy schedule.
Needs the app to perform reliably without interruptions, as she has limited time to troubleshoot issues.
Fitness Fanatic Felix:

Description: Felix is a fitness enthusiast who uses the Carb Manager App to track his macros, plan his workouts, and monitor his progress towards fitness goals. He's dedicated to optimizing his diet and exercise routines to achieve peak performance.
Testing Approach:
Explores advanced features related to fitness tracking, such as customizing workout plans and setting specific macronutrient goals.
Seeks detailed insights and analytics to monitor progress over time.
Values accuracy in tracking nutritional data and expects precise calculations for macros and calories.
Requires seamless integration with fitness devices and wearables to sync activity data effortlessly.
Looks for performance optimizations to ensure smooth usage during intense workout sessions.
Health-Conscious Hannah:

Description: Hannah is health-conscious and uses the Carb Manager App to support her dietary preferences, which include veganism and gluten-free eating. She's mindful of her food choices and seeks out products and recipes that align with her nutritional goals.
Testing Approach:
Explores features related to dietary restrictions and preferences, such as filtering recipes by dietary labels (e.g., vegan, gluten-free).
Checks for comprehensive nutritional information for various ingredients, including alternative food items commonly used in specialized diets.
Values inclusivity and expects the app to cater to diverse dietary needs beyond mainstream options.
Seeks community support and resources for sharing and discovering recipes suitable for her dietary lifestyle.
Requires clear labeling and categorization of foods to easily identify suitable options while shopping or meal planning.
Tech-Savvy Tyler:

Description: Tyler is a tech-savvy user who enjoys experimenting with new gadgets and software tools. He uses the Carb Manager App to track his nutrition and experiment with different dietary approaches, such as intermittent fasting and ketogenic diets.
Testing Approach:
Explores advanced settings and customization options to tailor the app experience to his specific preferences.
Tests compatibility with emerging technologies, such as voice assistants or smart kitchen appliances, to streamline data input and integration.
Looks for opportunities to leverage data export and API integrations for analyzing nutrition data in third-party tools or platforms.
Provides feedback on user interface enhancements and feature suggestions based on trends in technology and user experience.
Values continuous updates and optimizations to keep the app aligned with the latest advancements in nutrition science and technology.
Senior Sam:

Description: Sam is an older adult who uses the Carb Manager App to manage his dietary intake and support his health goals. He's relatively new to using digital tools and prefers simple, user-friendly interfaces that are easy to navigate.
Testing Approach:
Evaluates the app's accessibility features and usability for older users, such as font size options and clear navigation cues.
Tests the app's responsiveness to slower internet connections and older devices to ensure a smooth user experience.
Provides feedback on language clarity and instructional guidance to support users with limited technical knowledge.
Requires robust customer support resources, including clear documentation and easy access to assistance channels.
Values straightforward troubleshooting steps and intuitive error messages to address any issues encountered during app usage.

Now that we have persons ready let's focus on the PRD and test cases

PRD — Meal Plan Sharing

Here I will present a fictional PRD of a New Feature Implementation for the Carb Manager App

Feature Name: Meal Plan Sharing

Overview:
The Meal Plan Sharing feature enables users to create and share customized meal plans with friends,
family, or the Carb Manager community. This feature enhances user engagement and promotes
collaboration by allowing users to share their favorite recipes, meal ideas, and nutritional strategies.

Objective:
To provide Carb Manager App users with the ability to create, customize, and share meal plans seamlessly within the app, fostering a sense of community and enabling knowledge exchange among users.

Key Features:

Create Meal Plans:

Users can create personalized meal plans by selecting recipes from the Carb Manager database or adding custom meals and snacks.
Meal plans can be customized based on dietary preferences, nutritional goals, and meal frequency (e.g., daily, weekly).
Users have the option to specify portion sizes, serving quantities, and cooking instructions for each meal.
Share Meal Plans:

Users can share their meal plans with others via email, social media platforms, or direct links generated within the app.
Shared meal plans include a comprehensive overview of recipes, ingredients, nutritional information, and serving sizes.
Recipients can view shared meal plans without requiring a Carb Manager account, promoting accessibility and widespread adoption.
Discover Shared Meal Plans:

Users can explore and discover shared meal plans from other Carb Manager users within the app's community section.
Featured meal plans curated by the Carb Manager team highlight popular recipes, trending dietary trends, and user-generated content.
Users can search for specific meal plans based on keywords, dietary preferences, or nutritional criteria.
Collaborate and Comment:

Users can collaborate on shared meal plans by leaving comments, feedback, and suggestions for recipe modifications.
Comment threads facilitate discussions among users, allowing for recipe swaps, ingredient substitutions, and cooking tips.
Collaborative features encourage user engagement and foster a sense of community within the Carb Manager platform.
Use Cases:

Creating a Personalized Meal Plan:

Sarah wants to create a weekly meal plan tailored to her ketogenic diet. She selects keto-friendly recipes from the Carb Manager database and adds them to her meal plan, customizing serving sizes and portion quantities.
Sharing a Meal Plan with Friends:

John has discovered a delicious low-carb meal plan on Carb Manager and wants to share it with his friends. He generates a shareable link within the app and sends it to his friends via email and social media, inviting them to try out the recipes together.
Exploring Community Meal Plans:

Lisa is looking for inspiration for her upcoming meal prep session. She explores the community section of the Carb Manager app and discovers a variety of meal plans shared by other users. She finds a vegetarian meal plan that aligns with her dietary preferences and decides to give it a try.
Technical Requirements:

Backend Infrastructure:

Implement database architecture to store and retrieve user-generated meal plans, recipes, and associated metadata.
Develop APIs for creating, sharing, and accessing meal plans securely within the app.
User Interface Design:

Design intuitive user interfaces for creating, sharing, and exploring meal plans on both mobile and web platforms.
Ensure seamless integration with existing Carb Manager app features, maintaining visual consistency and brand identity.
Security and Privacy:

Implement authentication and authorization mechanisms to protect user data and ensure secure sharing of meal plans.
Adhere to data privacy regulations and guidelines to safeguard user information and maintain user trust.
Success Metrics:

User Engagement:

Increase in the number of active users interacting with the Meal Plan Sharing feature within the app.
Growth in user-generated content, including shared meal plans, comments, and interactions.
Community Growth:

Expansion of the Carb Manager community through increased user participation and collaboration.
Positive feedback and ratings from users regarding the usability and value of the Meal Plan Sharing feature.
Feature Adoption:

Percentage of users who create and share meal plans within the app, indicating adoption and user satisfaction.
Analysis of user retention rates and app usage patterns following the introduction of the new feature.
Conclusion:
The Meal Plan Sharing feature enhances the Carb Manager App's functionality by empowering users to create, share, and collaborate on personalized meal plans within the community. By facilitating knowledge exchange and fostering a sense of belonging, this feature contributes to a more engaging and rewarding user experience, aligning with Carb Manager's mission to simplify and optimize nutrition tracking for all users.

Requirements conversion into Test Cases

Converting a Product Requirements Document (PRD) into test cases involves a systematic approach to ensure that all functionalities and user scenarios are thoroughly tested. Here’s a step-by-step process:

  1. Review the PRD: Carefully read and understand the PRD to grasp the scope, features, and functionalities of the product.
  2. Identify Key Parts: Identify the key requirements, user stories, and acceptance criteria specified in the PRD.
  3. Break Down Requirements: Break down each requirement into smaller, more manageable pieces or user stories that can be tested independently.
  4. Create a Structure: Organize the requirements and user stories into a structured format, such as a mind map or a hierarchical list, to visualize relationships and dependencies.
  5. Identify Inputs and Outputs: Identify the inputs required for each functionality or user interaction (e.g., user inputs, system responses) and the expected outputs or outcomes (e.g., successful actions, error messages).
  6. Identify Preconditions: Determine any preconditions or assumptions necessary for executing the test cases (e.g., user authentication, system configuration).
  7. Generate Test Cases: Create detailed test cases for each requirement, user story, or functionality based on the identified inputs, outputs, and preconditions. Each test case should have a clear objective, steps to reproduce the scenario, expected results, and any additional information necessary for execution.
  8. Prioritize Test Cases: Prioritize test cases based on factors such as criticality, business impact, and frequency of use. Focus on testing high-priority functionalities first to mitigate risks effectively.
  9. Cover Edge Cases: Identify and include edge cases, boundary conditions, and exceptional scenarios that may occur during real-world usage. These test cases ensure that the product behaves correctly under various conditions and handles exceptions gracefully.

Traceability Matrix creation populated with personas

Test cases creation prompt (could be further enhanced i.e. with the information provided above)

create a listo of possible test cases based on Product Requirements Document

Response

TC001: Create Meal Plans

Description: Verify that users can successfully create personalized meal plans.
Testing Personas: Busy Bethany, Fitness Fanatic Felix, Health-Conscious Hannah

TC002: Share Meal Plans
Description: Validate the functionality to share meal plans with others.
Testing Personas: Busy Bethany, Fitness Fanatic Felix, Hipster Hilary

TC003: Discover Shared Meal Plans
Description: Ensure users can find and explore shared meal plans from the Carb Manager community.
Testing Personas: Hipster Hilary, International Ioana

TC004: Collaborate and Comment
Description: Test the ability for users to collaborate on shared meal plans by leaving comments and feedback.
Testing Personas: Hipster Hilary, Sneaky Shirley

TC005: Backend Infrastructure
Description: Verify the integrity and functionality of the backend infrastructure supporting meal plan creation and sharing.
Testing Personas: Sneaky Shirley, International Ioana

TC006: User Interface Design
Description: Evaluate the user interface design for creating, sharing, and discovering meal plans.
Testing Personas: Careful Claire, International Ioana

TC007: Security and Privacy
Description: Validate security measures and privacy protections related to sharing and accessing meal plans.
Testing Personas: Sneaky Shirley, Elder Elisabeth
These test cases cover various aspects of the Meal Plan Sharing feature, ensuring comprehensive testing across different functionalities and user scenarios.

Traceability Matrix creation step

create a Traceability Matrix table based on the information from 
the last response that would include testing personas created previously

Response:

Below is a simplified version of a traceability matrix based on the provided information:

| Requirement / User Story | Test Case ID | Testing Persona(s) |
|--------------------------------|--------------|--------------------------------|
| Create Meal Plans | TC001 | Busy Bethany, Fitness Fanatic Felix, Health-Conscious Hannah |
| Share Meal Plans | TC002 | Busy Bethany, Fitness Fanatic Felix, Hipster Hilary |
| Discover Shared Meal Plans | TC003 | Hipster Hilary, International Ioana |
| Collaborate and Comment | TC004 | Hipster Hilary, Sneaky Shirley |
| Backend Infrastructure | TC005 | Sneaky Shirley, International Ioana |
| User Interface Design | TC006 | Careful Claire, International Ioana |
| Security and Privacy | TC007 | Sneaky Shirley, Elder Elisabeth |

In this matrix:

  • Each row represents a specific requirement or user story.
  • Each column represents a test case.
  • The “Testing Persona(s)” column indicates which testing personas are relevant to each test case, based on the nature of the functionality being tested.

In this context, a follow-up prompt should be created to deficient between test cases execution in the context of presented personas

taking in to acount information given in the traceability matrix table 
present above create test cases for each person for a TC001

Response:

Certainly! Let's create test cases for TC001 (Create Meal Plans) tailored to each testing persona mentioned in the Traceability Matrix:

1. **Busy Bethany:**
- Test Case: BB_TC001
- Description:
- Ensure that Busy Bethany can quickly create a meal plan within a limited timeframe.
- Steps:
1. Log in to the Carb Manager App.
2. Navigate to the "Meal Plans" section.
3. Select the option to create a new meal plan.
4. Add pre-defined meal options or use quick-add functionality to populate the meal plan.
5. Save the meal plan and verify that it is created successfully.

2. **Fitness Fanatic Felix:**
- Test Case: FF_TC001
- Description:
- Validate that Fitness Fanatic Felix can customize a meal plan to meet his specific macronutrient goals.
- Steps:
1. Log in to the Carb Manager App.
2. Access the "Meal Plans" feature.
3. Customize the meal plan by adjusting serving sizes and portion quantities for each meal.
4. Set specific macronutrient goals (e.g., protein, carbs, fats) for the meal plan.
5. Save the customized meal plan and verify that the nutritional targets are reflected accurately.

3. **Health-Conscious Hannah:**
- Test Case: HC_TC001
- Description:
- Confirm that Health-Conscious Hannah can create a meal plan tailored to her vegan and gluten-free dietary preferences.
- Steps:
1. Log in to the Carb Manager App.
2. Navigate to the "Meal Plans" section.
3. Filter recipes by dietary labels (e.g., vegan, gluten-free) and select suitable options for the meal plan.
4. Add custom meals or snacks to complement the selected recipes.
5. Save the vegan and gluten-free meal plan and verify that it contains appropriate options.

4. **Hipster Hilary:**
- Test Case: HH_TC001
- Description:
- Ensure that Hipster Hilary can create a unique and eclectic meal plan using unconventional ingredients and units of measure.
- Steps:
1. Log in to the Carb Manager App.
2. Explore diverse recipe options, including less mainstream or trendy ingredients.
3. Experiment with alternative units of measure (e.g., grams, ounces, cups) for ingredient quantities.
4. Customize the meal plan with a mix of traditional and avant-garde dishes.
5. Save the eclectic meal plan and verify that all ingredients and measurements are accurately captured.

5. **International Ioana:**
- Test Case: II_TC001
- Description:
- Verify that International Ioana can create a meal plan while considering different time zones, keyboard layouts, and language preferences.
- Steps:
1. Log in to the Carb Manager App from a location outside local business hours.
2. Access the "Meal Plans" feature and create a meal plan using the app interface in multiple languages.
3. Test input compatibility with various keyboard layouts commonly used in different regions.
4. Save the meal plan and verify that it is accessible and functional across different time zones and language settings.

These test cases cover a range of user personas and scenarios, ensuring thorough testing of the Create Meal Plans functionality within the Carb Manager App.

Notice how test scenarios differ from each other based on persona characteristics, highlighting various user stories connected to different features and user needs.

Functionality (PRD) enhancement

Now we can iterate back to the PRD and as a follow-up question (bear in mind that here I am assuming that the chat context is persisted)

how can I further enhance given functionality

Response (notice that the names differ as the response pasted below is a PoC — it was part of a similar conversation with Gemini Pro v. February 13, 2024 and thus the names differ as well as personas characteristics):

## Enhancing the Meal Plan Sharing Functionality in Carb Manager:

Based on the personas and test cases, here are some ideas to further enhance the Meal Plan Sharing functionality in Carb Manager:

**For all users:**

* **Personalization:** Allow users to customize the meal plan sharing experience by setting privacy preferences for individual meals within a plan, choosing specific sharing groups, and customizing sharing messages.
* **Advanced filtering and search:** Implement advanced filtering options for shared plans based on dietary restrictions, allergies, ingredients, cooking time, and specific user-defined criteria.
* **Collaboration tools:** Enable real-time co-creation of meal plans with friends or family, including discussion threads, recipe suggestions, and collaborative editing.
* **Progress tracking and feedback:** Allow users to track the progress of shared meal plans, see how others are using them, and leave feedback and ratings for plans and individual recipes.
* **Seasonal and thematic plans:** Offer curated seasonal or thematic meal plan collections (e.g., holiday meals, budget-friendly options, international cuisines) alongside user-created options.
* **Integration with grocery shopping and meal prep tools:** Integrate with grocery shopping apps and meal prep tools to streamline the process of buying and preparing ingredients based on shared meal plans.

**For specific personas:**

* **Manager Maria:**
* Offer quick-start templates for common meal plans (e.g., "Busy Weeknight Meals") and one-click meal selection based on dietary preferences.
* Integrate with productivity tools and calendars to automatically schedule meal prep tasks based on the plan.
* **Careful Claire:**
* Provide detailed nutritional information and analysis for shared plans, including macronutrient breakdowns and personalized recommendations.
* Allow exporting plans with detailed recipe instructions and ingredient lists for easy printing or sharing.
* Offer educational resources and cooking tips within the app to support meal plan creation and execution.
* **Sneaky Shirley:**
* Implement robust security measures to prevent unauthorized access and plan manipulation.
* Allow reporting suspicious activity or inappropriate content within shared plans.
* **International Ioana:**
* Offer multi-language support and recipe options, adapting ingredient suggestions and measurements to local availability.
* Integrate with currency conversion tools to facilitate budget planning across different regions.

By incorporating these enhancements, Carb Manager can cater to a wider range of user needs and preferences, fostering a more engaging and collaborative community around meal planning and healthy eating.

Summary

The presented case study illustrates a proof of concept of AI-enhanced test case creation, leveraging both a Product Requirements Document (PRD) and a Person-Based testing approach. Persona-based testing is pivotal in validating software products by considering the behavior, preferences, and needs of diverse user personas. Through a traceability matrix, requirements are linked to test cases, aiding in efficient testing and comprehensive coverage tracking during development, thereby establishing a foundation for further detailed test case development. Notice that the case study presented above is just a PoC and my engagement with AI can be characterized as minimalistic — I used ChatGPT v. 3.5 and a bit of Gemini Pro v. February 13, 2024, both at this moment are freely available (other or upgraded models could be used to enhance the given responses), more sophisticated prompting techniques could be applied i.e. Few-Shot Prompting, Generated Knowledge Prompting — which should lead to further improvements of the presented approach of AI-aided Persona based Test Cases creation. On the other hand, you might have noticed that the preview image of this article (the one at the very top) faces of different people — it was generated by AI, a seemingly easy task was given “generate an image, picture faces of people of different age and gender” and although the outcome is close to what was expected, imperfections (one might even say glitches) are easily noticeable. A lesson to be learned here is to control, check, and change responses provided by the AI based on our knowledge — in other words, solving tasks with AI aid may only speed the process but will not allow us to fully automate it (not yet).

--

--