As AI continues to advance, experts are paying close attention to the rise of ‘agentic AI,” systems that can make decisions and take actions on behalf of users. These autonomous agents are being used to schedule appointments, pay bills and assist with medical research and software development. 

In Communications of the ACM, University of Michigan School of Information lecturer Merve Hickok, an expert on AI ethics and governance, says these systems may amplify problems like bias and misinformation. 

“Since we do not have a solution for value alignment or hallucinations, we should be worried about agentic AI systems based on language models,” Hickok says. “The same ethical concerns apply, with the additional element of more complexity. An individual agentic AI might contain bias or errors.”

Hickok adds that she does not think “we yet know the full extent of safeguards which may be necessary, or the effectiveness of the current methods we have” to control these systems.” 

RELATED 

Read “The Rise of the AI-Enabled Agentic Internet” at Communications of the ACM. 

Learn more about UMSI lecturer Merve Hickok by visiting her UMSI faculty profile.