Position: Generative Engine Optimization Creates Underexamined Risks, Governance Must Target Concentration, Disclosure, and Academic Blind Spots

The article discusses the emergence of Generative Engine Optimization (GEO) in the context of large language models (LLMs), highlighting risks associated with concentrated influence and undisclosed commercial biases in the evidence used by these models. It formalizes a GEO pipeline to identify optimization impacts and reveals academic-industry blind spots due to differences in evaluation practices. The authors advocate for enhanced governance measures, including answer-level oversight, precise disclosures, and black-box auditing to mitigate these risks and improve the reliability of LLM outputs.