Expert Forum Seminar 2
Date: 29 september 2026
Location: Chalmers, Gothenburg
Free of charge: Membership in Swedsoft or Expert Learning Lab required.
Language: The seminar will be held in English.
Expert Forum 2 – Back to Basics: Why Testing and Verification Are Even More Important When Using Agentic A
Following the engaging discussions at our Expert Forum in May, where we explored the opportunities and challenges of Agentic AI in software development, one message stood out clearly: while AI has the potential to dramatically accelerate development, it does not reduce the need for testing and verification. If anything, it makes them more important than ever and junior developers needs to understand and go back to basics as producers of software.
We therefore invite software developers, testers, quality engineers, researchers, and academics to join us for Expert Forum 2 – Back to Basics, where we will explore how testing, verification, and quality assurance must evolve to keep pace with increasingly autonomous AI-driven development practices.
Agentic AI can generate code, design solutions, and automate complex development tasks at unprecedented speed. However, code that appears correct may still contain logical flaws, incorrect assumptions, security vulnerabilities, or integration issues that only emerge in real-world environments. As AI agents become more autonomous, small errors can quickly propagate across systems if they are not identified through systematic testing and verification.
Another key challenge is integration with existing systems. Most software exists within a complex ecosystem of applications, databases, APIs, and business processes. An AI-generated solution may perform well in isolation while creating unexpected issues when integrated with legacy systems or established workflows. This makes integration testing, regression testing, and system verification critical to ensuring reliable operation across the entire technology landscape.
During this forum, we will discuss questions such as:
- How should testing practices evolve in an era of Agentic AI?
- What new risks are introduced by autonomous software development?
- How can we ensure that AI-generated software meets business and user requirements?
- What role do integration, regression, and security testing play in AI-assisted development?
- How can industry and academia collaborate to build trust in AI-generated software?
The promise of Agentic AI is significant, but organizations will only realize its full value if they can trust its output. That trust is built through rigorous testing, verification, validation, and quality assurance processes. The faster and more autonomous software development becomes, the more important these foundations are.
Join us for an interactive discussion where practitioners and researchers come together to share experiences, challenges, and ideas on how to balance innovation with software quality in the age of Agentic AI.
Who should attend?
Software developers, testers, quality engineers, software architects, researchers, academics, and anyone interested in the future of trustworthy AI-assisted software development.
We look forward to seeing you there!