AI Agents in the workplace raise new ethics and job security concerns

Workforce disruption: There's much speculation that employees face a shaky future as AI becomes increasingly adept at filling their roles. Mistry narrows in on Open AI's Operator as a benchmark for agents, and potential threat. “A solution like Operator can potentially impact any job that performs non-innovative functions,” says Mistry. “For instance, millions of people who are employed in call centers can be replaced by AI agents since the majority of them cycle through repetitive scripts and tasks. Call centers could run 24/7 with much less liability. Since these AI agents operate using the chain-of-thought method, they can eventually be used in applications that include reasoning, such as accounting and HR.”

Government's role: The threat to employment stability demands action. “AI agents are on track to replace a large number of people and remove them from jobs that they rely on to survive," Mistry warns. "Thus, governments need to be involved here and put checks in place to make sure people’s livelihoods—and by extension the economy—are not completely uprooted."

Current reality: Despite the fear around job replacement and the need for protections to be put in place, Mistry points to Operator’s function of giving 'control' back to a human when it faces a challenge that it cannot tackle. In his eyes, "having some form of human involvement will always be necessary." Without such oversight, he says, AI agent systems risk falling into "an endless cycle of committing mistakes." AI's current susceptibility to errors and its need for humans in the loop provides some protections to people—at least for the time being.