The software testing trends of 2024 came down to three forces: AI tools moving into the test process, the EU Accessibility Act driving accessibility testing, and the Cyber Resilience Act setting a mandatory framework for security testing. AI does not replace human testing experience. It is a tool that makes good testers stronger and lets poor ones produce bad results faster.
Key Takeaways
- AI-supported test tools are standard now: hardly any serious test tool comes without built-in AI features for test case design, GUI automation or log analysis.
- A good tester gets better results faster with AI, a poor tester gets worse results even faster.
- Laws such as the EU Accessibility Act, the Cyber Resilience Act and the EU AI Act are the strongest drivers for building quality checks into development earlier.
- Low-code and no-code test automation lets subject matter experts without programming skills specify and run tests themselves, and it will gain further weight in 2025.
Software Testing Trends 2024: AI Has Arrived, the Experimental Phase Is Over
Among the software testing trends of 2024, one stood above the rest: artificial intelligence went from hype topic to a fixed part of everyday test work. Early in the year, many people opened a language model, typed a single sentence and hoped for complete test cases and test data. Today they work more soberly. The initial excitement has given way to a critical look, both at testing AI and at testing with AI.
Florian Fieber sums it up like this: AI was the overarching topic at almost every conference, from events with their own AI formats such as the SIQ Info Days to Testing United, which was devoted to the topic entirely. The key lesson of the year is that when it comes to testing with AI, not everything that glitters is gold.
People stay in the loop. AI amplifies existing skills, it doesn’t replace them. A good tester working with AI delivers better results and works faster. A poor tester working with AI produces worse results, only faster.
“There’s no way we can take people out of this, not our skills as testers and not our experience. That is fundamental, and it stays.”
(Florian Fieber)
How AI Becomes an Everyday Testing Tool
AI is turning into a commodity, a tool you simply use in the test process. Right now, nearly every individual tester uses a language model for their own tasks, but it is rarely built firmly into the test and development process.
When it comes to testing with AI, adoption is moving into the tools. There are hardly any test tools left without built-in AI, whether for test case design, scripting, coming up with test ideas, recognizing UI elements in GUI automation or finding patterns in log files and error logs.
AI assistants in particular have become easy to use. With APIs and platforms such as OpenAI’s, you can build your own assistant for a specific use case, for example one that takes requirements as input and helps develop test ideas and acceptance criteria from them.
It makes sense to build up a collection of such assistants, each supporting one test activity. Tools like these are gradually finding their way into many places.
AI Adoption Is Uneven
AI is not spreading through testing evenly. It is not an either-or development, and it doesn’t move at the same speed across test activities, domains or organizations.
Some teams go all in, while other areas are still far from it. That pattern is familiar from earlier methods and tools that were introduced over the years.
There is a gradient within the test activities, too. In test execution, test case design and evaluation, AI will become very strong. In test planning, monitoring and control, the level of autonomy will stay lower for a long time.
Why Laws Are the Strongest Drivers in Testing Right Now
In the broadest sense, legislation drives testing topics. Codified, verifiable product requirements are ideal from a testing perspective, because they draw attention and push the topics forward.
Three regulations are acting as drivers at the moment:
| Regulation | Topic in Testing |
|---|---|
| EU Accessibility Act | Accessibility testing, access to products |
| Cyber Resilience Act | Cybersecurity, organizational and process evidence |
| EU AI Act | AI governance, testing and assessing AI systems |
Next to AI, accessibility testing was the second big conference topic of the year, driven by the Accessibility Act coming into force soon. Cybersecurity and the Cyber Resilience Act also created a great need for information, and that need has not been met yet.
These requirements go beyond testing alone. Cybersecurity also covers organizational and process aspects as well as questions of evidence that reach past classic testing.
Non-Functional Quality Belongs in the Design, Not at the End
Security, accessibility and similar qualities have to be considered and built in from the start. You can’t successfully test them into a finished product after the fact.
The common reflex looks different. First the software is finished, then comes the performance test, the security test or the accessibility check, without anyone having dealt with the topic in the months before.
The challenge is to build the relevant tests into the pipelines and into continuous integration. Teams that check accessibility only at the end find open issues and then pull the checks forward step by step. It makes more sense to reach into requirements engineering from the very beginning, so the qualities are built in and testing only has to confirm them.
Agile Is Unfinished Business, and Quality Is Its Lever
Many organizations are further from working in a truly agile way than they think. Agile transformation, scaling and adoption still have a long way to go, and testing plays a major part in that.
Agility forces quality. If you work in short iterations and don’t invest in testing and quality, the software will blow up in your face after a few sprints. Years ago, that insight was uncomfortable. Today it is a strong driver for testing in the right places.
In this setting, DevOps and quality engineering are gaining weight. Development and testing are becoming more closely interlinked, and testing is moving earlier into the development process.
How Low-Code and No-Code Open Up Testing
Test automation keeps gaining importance, and low-code and no-code are moving fast. The goal is to decouple the technical side from the business side.
The principle isn’t new. Keyword-driven testing has long aimed to abstract away from the test object and the technology underneath, so that nobody has to write their own test scripts anymore.
For you, this means business users and domain experts can specify and run tests without getting deep into the technology. Enabling non-technical people this way brings more domain knowledge into test automation.
International Standardization Thrives on Different Perspectives
Standardization in testing makes sense because it creates a common denominator, a basic framework and shared terminology. The market and the participants keep reflecting that need back.
Looking across borders changes your own view. How certification schemes, and the testing discipline as a whole, are perceived differs considerably between Germany, Austria, Switzerland and other countries. Knowing these perspectives is important, but it takes time and calls for compromise.
The work swings between two extremes. On one side there is the risk of overregulation and overstandardization, on the other the risk of developing something too quickly.
At the ISTQB, a dedicated working group structures testing in specific domains. The Certified Tester portfolio already includes modules for automotive testing and for gaming and gambling, and there is room for more. Syllabi on DevOps, on testing with AI and on accessibility testing are likely to follow sooner or later.
Frequently Asked Questions
Will artificial intelligence replace experienced testers?
No. AI enhances existing skills; it does not replace them. A good tester delivers better results and works faster with AI, while a poor tester produces worse results with AI, just faster. Testing experience and judgment remain integral to the process. The expectation that a language model would generate complete test cases and test data from a single sentence has not been met.
In which testing activities does AI provide the greatest benefit?
AI excels in test execution, test case creation, and evaluation, such as pattern recognition in log files and error logs. In test planning, monitoring, and control, however, the degree of autonomy remains much lower. Adoption varies widely anyway: Some teams go all-in, while other areas are still a long way from doing so.
Why can’t accessibility simply be tested at the end of development?
Features such as accessibility and security must be built in; they cannot be tested into a finished product after the fact. The common approach is different: First, the software is completed, followed by performance, security, or accessibility testing. It makes more sense to address these qualities from the start in requirements engineering and integrate the tests into pipelines and continuous integration.
Which legal requirements are driving testing priorities?
The EU Accessibility Act, the Cyber Resilience Act, and the EU AI Act were the strongest drivers in 2024. From a testing perspective, codified, verified product requirements are beneficial because they draw attention to the issue. The requirements extend beyond pure testing: cybersecurity also encompasses organizational and procedural aspects as well as compliance issues.
Do subject matter experts need programming skills to automate tests?
No. Low-code and no-code approaches decouple technology from domain expertise, allowing business users and domain experts to specify and execute tests without delving deeply into the technical details. This principle is not new: Keyword-Driven Testing has long aimed to abstract away from the test object and the underlying technology. The benefit lies in incorporating more domain knowledge into automation.
What is the connection between agile working methods and testing?
Agility forces quality. Anyone working in short iterations who doesn’t rely on testing will see the software blow up in their face after just a few sprints. Many organizations are further from a consistently agile way of working than they realize. In this environment, DevOps and quality engineering are gaining importance, and development and testing are becoming more closely intertwined.
Are there certifications for testing in specific industries?
Yes. In 2024, a dedicated working group at the ISTQB structured testing in specific domains; the Certified Tester portfolio included modules for automotive testing as well as for gaming and gambling. Other topics were on the horizon, such as curricula for DevOps, AI-based testing, and accessibility testing. Standardization creates a common ground and consistent terminology.
