Overview
Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore the critical challenges facing modern software development in this 55-minute keynote conference talk that examines how advanced development tools and generative AI are creating new risks in software reliability. Discover how the evolution from predictable, traceable code to AI-assisted development has fundamentally changed the nature of software creation, making it easier than ever to produce systems that appear to work correctly but may harbor hidden flaws. Learn about the growing problem of "probably-working software" - applications that seem functional on the surface but lack the predictability and verifiability of traditional code. Understand the implications of lowered barriers to entry through generative AI tools that can create functional software from simple prompts, while potentially introducing opacity and brittleness into systems. Examine strategies for maintaining code quality and system reliability in an era where developers are increasingly removed from direct code control. Gain insights into determining when "good enough" truly meets quality standards and how to recognize when human intervention is necessary in AI-assisted development processes. Address the ethical considerations and testing challenges that arise when working with tools that make software development faster but potentially less reliable.
Syllabus
Keynote: The dangers of probably-working software - Damian Brady - NDC London 2026
Taught by
NDC Conferences