AI in Software Development and Its Potential Influence on Accessibility Compliance

As artificial intelligence (AI) becomes increasingly integral to various sectors, its potential to enhance accessibility in digital services is of growing interest. This paper explores the influence of AI systems (AIS) on accessibility compliance, specifically within the context of Norway's national e-health service provider Norsk helsenett (NHN). Through a qualitative analysis, we assess how AISs are deployed in frontend programming, user interface design, and accessibility testing at NHN. Our study reveals that while AI has the potential to improve accessibility, it also presents challenges that require robust domain knowledge and effective human oversight. We explore the benefits and limitations using ChatGPT in accessibility testing, GitHub Copilot for generating accessible code, the implications of AI driven design tools, as well as the key differences between developing websites with existing component libraries and developing a simpler website from scratch. This paper aims to contribute to the ongoing discourse on AI and accessibility, offering insights and recommendations for future research and practical applications.