AI has the potential to change every aspect of our daily lives. Still, it also has the potential to cause harm, especially to communities that are already facing inequity and discrimination. This risk of harm means that we need ethical guidelines to regulate AI. However, the rapid advancement of AI and the inflexibility of government regulation have made creating such guidelines challenging.

Researchers at Texas A&M University School of Public Health are creating a new model for how ethical guidelines and enforcement should be handled as artificial intelligence (AI) technology advances. This model, called Copyleft AI with Trusted Enforcement (CAITE), is designed to prevent the potential harm that AI could cause while allowing technological advancements to continue.

The CAITE model combines two approaches to managing intellectual property rights: copyleft licensing and the patent-troll model. Copyleft licensing allows sharing intellectual property under certain conditions, such as attribution and noncommercial use. The patent-troll model uses enforcement rights to ensure compliance.

The CAITE model is built on an ethical use license that restricts unethical AI uses and requires users to follow a code of conduct. It uses a copyleft approach to ensure that developers who create derivative models and data must use the same license terms as the parent works. The enforcement rights for all these ethical use licenses would pool in a single organization, called a CAITE host, which would act as a quasi-government regulator of AI.

The CAITE model is designed to be fast and flexible like industry but with the enforcement power of a traditional government regulator. The model also allows for leniency policies and incentives that promote self-reporting and trust in oversight.

The authors note that the model will require significant community buy-in and government incentives to be implemented. However, industry is likely to prefer the more flexible CAITE framework to the stringent and slow-to-adapt regulations that governments could eventually impose.