AI in Testing 2026: Tool, Colleague, Test Object
AI writes tests, explores apps and generates test data. And it becomes a test object itself. Five fields that really matter for testing with GenAI in 2026.
AI & Software Quality
Artificial intelligence changes software quality in two directions: teams need to evaluate AI-based systems with probabilistic outcomes, while generative AI can also support analysis, test design, and automation. This topic separates these perspectives and connects technical opportunities with realistic quality risks.
The articles and podcast episodes cover areas such as testing large language models, data and model quality, suitable evaluation approaches, and the responsible use of AI in testing. They explore where AI provides productive support, where human judgement remains essential, and how experiments can become traceable quality practices.
Foundations
Articles
Faster Horses
AI is doing our old job faster, like a horse with more horsepower. Which skill the next abstraction layer devalues, and which one it rewards.
Between Pride and Throwaway Software
Taking over AI output builds no relationship with the result. Why ownership grows from working through things, not from generating them.
The Great Compression
Kamchatka 2008 with books instead of a ChatGPT itinerary: what the double compression of time and work does to us, and why deliberate brakes help.
AI: Are We Solving the Problems That Matter?
AI generates movies and apps while teams drown in legacy code. Why we should use AI for the problems that actually hurt, not for demos.
Relaxed, but Mindful
AI noise from every direction: why total refusal rarely works, and how small steps and the Gartner Hype Cycle help you stay relaxed but mindful.
Of Tools and Trailblazers
The future of testing lies not in better tools but in humanity. Why testers are the interpreters between AI, code, and the people behind it.
The future of testing
Explore the future of software testing with AI in software testing and essential soft skills for achieving better quality and user satisfaction.
When AI is developing
Explore how AI in software development is reshaping the landscape, making human-centric programming languages potentially obsolete.
Stoicism - decision making in the world of AI
Explore Stoicism in the AI era: reduce decision stress and navigate life with stoic principles for a focused mindset. Empower your decision making.
Focus on people
In 2025, agile coaching techniques empower software developers to thrive alongside AI, enhancing creativity and talent in development processes.
(Test)data radical cure
Explore how effective test data management can drive software quality. Discover strategies for overcoming testing challenges in 2025.
Testing of and with AI
Discover how AI in software testing can enhance your quality and efficiency, transforming your testing approaches in 2025.
Software Testing and AI - Man Against Machine
AI is transforming Software Testing and AI strategies. Discover how innovative tools enhance testers' efficiency in 2025.
Software testing in the future - Interview with Nico Liedl
In 2025, discover the future of software testing with agile quality management integrated with AI and new methodologies.
Software testing in the future - Interview with Michael Mlynarski
Explore the future of software testing with AI in software testing techniques revolutionizing quality assurance strategies.
Software Testing of and with AI
Explore Software Testing with AI and discover the synergies, opportunities, and challenges of this innovative technology in software development.
Podcast Episodes
AI as a Tester: How LLM-as-Judge Works
Automating LLM tests, even though no result is guaranteed? It’s possible—if you redefine determinism and use LLMs as judges.
Boilerplate in Seconds: AI Handles Setup, Engineers Handle Logic
Small, focused AI functions in testing outperform big ambitions. Keyword docs, reviews, defect translation: what actually works and where LLMs still fall short.
Vibecoding and the Lack of System Understanding in Practice
The middle tier of developers is disappearing; those who remain need a broad understanding of the system and real depth, rather than just routine “vibe coding.”
Automatically Reviewing Documentation with AI Personas
AI Personas as Documentation Testers: What a Simulated Junior Developer with 1.8 Million Tokens Reveals About Where Documentation Really Falls Short.
ChatGPT Use Cut Student Cognitive Capacity, Study Finds
Students who used ChatGPT lost 47% of their cognitive capacity. What that means for how we work, hire, and build software teams in the AI age.
What AI Really Does to Trust and Team Dynamics
AI makes team communication more objective, and that is not the same as progress. What trust and empathy lose, and how teams push back.
Critical Thinking: The Skill AI Cannot Replace in Testing
Cutting QA teams doesn't eliminate bugs, it just makes users find them first. Here's why communication and critical thinking still matter most.
Strategy First: How AI Enters Regulated Medical Labs
AI in a regulated medical lab sounds like a compliance nightmare. Here is how a strategy-first approach made it work without breaking the rules.
When generative AI violates your own values
Generative AI cannot be used ethically as long as training data is used without consent and billionaires control the models.
Why Traditional Testing Fails for AI Systems
Testing a chatbot breaks every rule traditional testing relies on: same input, wildly different outputs, and bugs that live outside the code.
Why Testers Are Safe Despite AI Hype
AI won't replace testers, but it will shift what they do. Where it saves real effort, where it still needs human control, and why starting now matters.
Why agentic engineering changes everything
If you have bad principles and use AI, you will get worse faster. Why agentic engineering needs more than code generation.
Security tests for AI systems
AI systems can be tricked into revealing protected data through clever prompts. Where the points of attack lie and what OWASP recommends.
Building Trust with AI Agents
Trusting an AI agent works the same way as trusting a colleague: you need clear communication, checks, and a system that catches what the model gets wrong.
Why AI cannot do cause-and-effect - and QFD helps
Why can't LLMs do cause-and-effect? And what does Quality Function Deployment have to do with it? The answer changes how you prioritize testing.
AI-generated test cases in a regulated environment
AI-generated test cases in the medical technology environment: how a RAG system remains regulatory clean without tool validation.
Software engineering of tomorrow
AI-generated code sounds tempting, but who pays the bill? Why software engineering needs more quality control, not less.
Critical Thinking in Software Testing
AI writes the code, AI writes the tests — but who checks if any of it is actually right? Critical thinking is the skill testers cannot afford to outsource.
AI-supported test case determination
Specialist departments hardly test at all because test knowledge is missing in the teams. An AI assistant delivers cases to ISO standards.
Ask Me Anything about AI, automation and Shift Left
176 episodes, 122,000 downloads, flaky tests, AI fears and a microphone faux pas: what the year really left behind.
Year-End Review: AI and Accessibility
AI won't replace testers, but it will reward those who use it well. Here's where the real productivity gains are in 2026.
Software testing Christmas chat 2025
AI doesn't make bad processes better, it catches up with them faster. What testers really need in 2026: Gut feeling, basic knowledge and community.
Legacy apps automated
From two weeks of manual testing to three hours: how an AI-based solution automates legacy apps without element IDs.
Man vs. machine: Who judges more fairly?
AI systems have bias, but the real problem is that people adopt it without realizing it. What this means for the use of AI in companies.
More quality in requirements with AI
Vague requirements are the most expensive problem in testing. How AI really helps when determining, formulating and checking requirements.
Testing with Natural Intelligence
AI takes over testing? What it can't do: genuine curiosity, intuition and thinking outside the box. Why human tester thinking remains.
Results of the 2024 software testing survey
A third of respondents already use AI for coding, yet regression testing lacks automation. What the 2024 software testing survey shows.
Static analysis with AI
Static analysis throws up thousands of findings and AI fixes two thirds reliably. What that means for old code bases and where it stops.
Still Coding or Just Prompting?
Most programmers using AI tools will create legacy code faster, not better. Here's what actually separates quality developers in 2034.
AI Agents & the Future of Testing
Trusting AI agents with your business works like hiring a new colleague: you need selection, probation, and continuous performance checks built in.
Legacy modernization
Who modernizes legacy code when there are no experts? RAG-based AI draws knowledge directly from legacy code - and makes subject matter experts replaceable.
AI testing and certification
Testing AI without structure is like testing without scale. How a three-dimensional matrix brings order to the chaos of AI testing dimensions.
Analyzing and improving requirements
Requirements often have more gaps than expected, and DEFOSPAM uncovers them systematically. AI can support exactly this analysis step well.
Test description for AI capabilities
Making AI systems testable: How capabilities, quality criteria and structured test descriptions turn abstract standards into concrete testing approaches.
Test design with AI
Many people know test design methods, but hardly anyone uses them. Why this is the case and how AI closes this gap without making you stop thinking.
Review 2024 and trends 2025
AI is becoming the standard tool in testing, accessibility is becoming mandatory and test automation continues to gain momentum. What really matters in 2025.
Trends in testing
AI dominates programming and testing, but lags behind in requirements engineering. Where Scrum really stands today and what that means for software quality.
GenAI in test automation
AI-generated test code compiles cleanly but very often tests the wrong thing entirely. Writing unit tests yourself is the smarter route.
How AI can support accessibility
Accessibility tests today end up in a jumble of Word, Excel and browser tabs. How an integrated test environment can change this and how AI can help.
AI testing - a checklist
Testing AI sounds complex but often boils down to one question: deterministic or not? What that means for your whole test strategy now.
AI, test automation and skills
AI testing, test data, accessibility: which trends testers should really have on their radar and why the right tool often doesn't solve the actual problem.
Software engineering in the year 2034
AI doesn't make developers more productive, it makes them faster at creating legacy code. What this means for quality and jobs until 2034.
Practical tester training with AI
Software testing often shows a gap between theory and daily practice in projects. AI-supported certification aims to close exactly that.
Acceptance test-driven LLM development
LLMs testing like a pro: Acceptance Test Driven Development meets fine-tuning - creating a measurable quality process for AI systems.
ChatGPT for testing
Using ChatGPT for test cases, test data and exploratory ideas: What really works, where caution is advised and why prompting matters.
Quality assurance of AI
Testing AI means rethinking: no clear test oracle, statistical quality instead of pass/fail. What this means in concrete terms and which methods really help.
Fair, good AI?
Fair AI sounds good, but what does it mean in measurable terms? Why fairness measures can contradict each other and how a structured assurance case can help.
Testing audio AI
Testing audio AI without being able to look into the code: Why golden test sets, training data tracking and ChatGPT play a role in this.
Quality from and with Prompt Engineering
AI can generate test cases, but can you rely on their correctness? That remains risky. Which prompt patterns really help and where the limits lie.
Will AI replace the test engineer?
From carpenter to agile engineering coach: How a career change and a book about system testing sparked an entire career in testing.
AI revolution in test automation
200 failed tests, but only seven real causes: How machine learning radically simplifies testing operations without rationalizing away roles.