Uveal melanoma is an incredibly common, but delicate form of cancer due to its intraocular nature. Incorporating a higher level of precision into intraocular cancers can aid in mitigating additional negative side effects of radiation for patients with salvageable vision. Artificial intelligence offers a new opportunity to modify the doses of radiation given to uveal melanoma patients depending on radiogenomic data. Machine learning algorithms have shown particular success in classifying tumors based on image biomarkers and multi-omics data to determine their sensitivity to radiation. Incorporating this higher level of precision into intraocular cancers can aid in mitigating additional negative side effects of radiation for patients with salvageable vision. However, there is significant concern regarding the diversity of the data which this AI is being trained on. Biased healthcare data, which reflects the medical inequalities in modern medicine, can leave minorities overlooked and underrepresented in life and death scenarios such as cancer diagnosis and treatment. Especially when training these new systems which determine the future of medicine, it is vital that the medical community prioritizes diverse and accurate data in the machine learning process. To maintain this ideal, the AI systems must be under consistent human oversight from both physicians and auditors. Overall, the introduction of AI machine learning into uveal melanoma treatment can vastly improve the quality and precision of care for patients, but only if it is borne with a dedication to diversifying data to ensure that each individual is given access to proper and thoughtful medical care.