Advancing lawful AI through 'compliance by design'

As AI systems increasingly shape all aspects of society and become progressively opaque, it is difficult for regulators and law enforcers to monitor their effect, identify perpetrators of unlawful AI-enabled and AI-assisted acts and gather the necessary evidence to undertake legal proceedings. In such a context, the purpose of this article is to present a new ‘compliance by design’ framework. This proposed framework involves embedding technical controls, protocols and tools within the design of AI systems to mandate compliance and advance the lawful application of AI. Such a framework can support compliance with a range of laws, including consumer, administrative, digital privacy and anti-discrimination laws. Enacting AI laws is the easy part. Ensuring that such laws can be effectively enforced is a greater challenge. While the European Union has enacted an AI Act and governments around the world have released proposals for AI legislation, little has been written about how such laws can be enforced. It is suggested that the strategy of compliance by design offers a logistically feasible technical solution to a critical legal problem of our time — the lawful deployment of AI.