Let me share with you the relationship between AI momentum in testing with automotive industry in the beginning of the1900 century.
Introduction
Artificial intelligence (AI) is revolutionizing the field of software testing, bringing unprecedented efficiency and accuracy to the process. At InnoWave, our NEXT program is dedicated to delivering value through AI solutions, and my involvement in implementing those innovations has been both fast-paced and rewarding.
This article provides an in-depth overview of AI’s impact on modern testing practices.
What Do People Want from AI in Testing?
According to the World Quality Report 2023, the top priorities for AI in testing are “Higher Productivity” and “More Velocity”.
This mirrors the historical analogy of Henry Ford’s clients wanting faster horses, highlighting that we are at the stage of people not envisioning a revolutionary change but (significant) incremental improvements.
What the Industry Says
If we search different sources of information, all of them are aligned with “more speed”. Advocating for generative AI to accelerate testing processes.
Even through testing tools, AI is transforming the way we approach testing. The table below shows important features provided by some testing tools and those features act as enablers to increase speed in current activities.
A quite different example is ISTQB® training program in AI Testing where we can see a dedicated chapter for “Using AI for Testing”, which have these topics:
- Using AI to Test Case Generation
- Using AI for the Optimization of Regression Test Suites
- Using AI to Analysis Defect Reports
- Using AI for Defect Prediction
- Using AI for Testing User Interfaces
The first three topics are aligned with “more speed” under the same paradigm of the last 20 years. The last two topics use AI to create a new paradigm.
Defect Prediction uses AI to estimate the amount of future defect based on the history.
Testing User Interfaces is under the concept of Visual Testing. Visual testing aims to verify that the application’s visual elements, like colors, images, fonts, and layouts, are displayed correctly and consistently across different devices, operating systems, and browsers. And using AI to do this automatically is clearly a game changer.
Nevertheless, the movement from “faster horses” to a new industry (automotive industry) will take some time to happen, but in the current present, we already have some topics touching that future. It is like to have a car with maximum speed of 10 or 20 Kms/h.
Nevertheless, the movement from “faster horses” to a new industry (automotive industry) will take some time to happen, but in the current present, we already have some topics touching that future. It is like to have a car with maximum speed of 10 or 20 Kms/h.
Testing Process and Tools
Other interesting point of analysis is to look at the testing process and the AI features in testing tools. You will see (again) a focus on accelerating testing activities using the same process that we use today.
Key AI Features in Testing Tools
Through short research, we can see the key AI Features in Testing Tools summarized as:
- Interpretation: AI interprets specifications, requirements, screenshots, test cases, and anomaly reports.
- Generation: AI generates test conditions, test cases, and test code.
- Summarization: AI summarizes progress, anomaly, and coverage reports.
- Evaluation/Correction: AI evaluates and corrects test coverage and test code.
The picture below shows the link between Key AI Features and the (usual) Testing Process:
The Evolving Role of Testers
We can see in the picture above that AI accelerates all activities before and after “Test Execution”. So, focus is using the same (traditional) engine for the execution activity.
My vision is AI will replace traditional engine with a new engine (similar to Visual Testing explained above) that validates software with minimal human interaction. The “Test Execution” box (in the picture above) highlights this trend… a new AI Test Engine. An engine smart enough that requires no scripting or a minimum number of lines of code. Not ready to be used now, but I believe within few years we will have this AI Test Engine, that receives my desire (my prompt), and it does all the testing activities based on my feature description.
However, the tester’s role remains crucial.
As we know, Agile practices support more on communication than documentation. And AI Engine requires documentation (that don’t exist in many cases).
Therefore, testers’ roles will be to provide quality data for efficient AI engines. Testers collect data from daily meetings, and they store that data into a comprehensive requirements repository.
Therefore, testers’ roles will be to provide quality data for efficient AI engines. Testers collect data from daily meetings, and they store that data into a comprehensive requirements repository.
This important activity ensures data quality that is mandatory for an efficient AI Engine. Next week, my colleague Miguel Figueiredo, will provide a new article where you can see a roadmap to ensure Data Quality in your organization. Stay tuned!
With this approach, you can imagine that testers will accumulate one more task and, therefore, they will have an increased work overload. But this won’t be the case.
Why? Because the comprehensive repository will enable automatic test case generation, transforming the testing process in a full automated pipeline and reducing manual efforts.
Conclusion
Currently, AI is essential for accelerating activities within the existing testing process, improving performance akin to “faster horses”.
In the near future, I believe we will witness a paradigm shift, where machines test other machines with minimal human interaction, heralding the rise of the “car” industry in software testing execution.
At InnoWave, our NEXT program leverages Agentic AI to revolutionize QA and the SDLC, delivering unparalleled agility, faster time to market, and significant cost reductions.
At InnoWave, our NEXT program leverages Agentic AI to revolutionize QA and the SDLC, delivering unparalleled agility, faster time to market, and significant cost reductions.
By integrating AI-driven solutions, we drastically minimize human errors and streamline testing processes, ensuring high-quality software delivery. Our expertise in implementing these innovations positions us as a leader in the industry, ready to transform your testing practices and drive your business forward.
