AI Risks Put on the Table in Malta's Gaming Sector
A survey by the Malta Gaming Authority shows that AI use is spreading across the sector, but it is concentrated mainly in areas with little direct contact with players. It also reveals serious gaps in AI requirements, documentation, data management and transparency.
According to a survey by the Malta Gaming Authority (MGA), artificial intelligence has taken hold in Malta's gaming sector. However, its use is concentrated in areas where its direct impact on players is limited. Customer support and operational analytics are the most widely adopted areas, while automated KYC checks and payment applications are rarely seen.
The findings were announced alongside the 48-page Artificial Intelligence Gaming Requirement published on 18 September 2026 by the MGA and the Malta Digital Innovation Authority (MDIA). Drawn up after consultations with license holders and technology experts, the framework offers voluntary guidance on AI governance. The requirement provides direction in six areas: transparency, fairness, sustainability, data protection, human accountability and technical reliability.
The research found gaps in how companies document and audit their technologies. Although most respondents consider their technical infrastructure adequate, fewer companies have a formal AI strategy or comprehensive risk management procedures. Policies on access, ownership and retention of data are mostly in place, but only a limited number of license holders can present a data lineage record showing the origin of the information used in their AI models.
Bias testing, impact assessments and staff training are also not sufficiently developed. Few participants reported that they have fully established AI risk assessment and incident response arrangements. Transparency concerns regarding external technology providers also stand out. Of the provider participants, only one reported that it provides customers with a document explaining how its systems work. This can make it harder for operators to explain third-party fraud detection or responsible gaming services.
The requirement classifies systems according to their potential impact. Applications affecting account restrictions, eligibility, fraud monitoring and player protection are subject to closer scrutiny. Chatbots must identify themselves as AI at the start of an interaction, and conversations involving vulnerability or gambling harm must be passed on to human staff. The requirement also warns against AI washing, where companies exaggerate their technology's capabilities or present conventional software as artificial intelligence.
Comments (0)
Be the first to comment.
Log in to comment.