At a Cybercab event this week, Tesla unveiled its Optimus robots, capturing attention with their ability to interact, serve drinks, and entertain guests with dance performances.

The event was filled with visuals showcasing the robots engaging with attendees — even answering questions — that hinted at remarkable technological advancement in a short time. Yet questions about the robots’ autonomy surfaced quickly.

The robots’ interactions were touted as a significant step forward in real-time response to verbal and visual cues, potentially marking a breakthrough in AI technology. Yet accounts from those present painted a slightly different picture.

A tech enthusiast, Robert Scoble, shared insights from the event, suggesting that human operators were “remote assisting” the robots, Verge reported.

Further clarification came when an engineer explained that the AI primarily facilitated walking functions, as reported by Electrek. A Morgan Stanley analyst, Adam Jonas, also commented on the use of “tele-ops (human intervention),” reinforcing the notion that the robots were not operating entirely autonomously.

Observations during the event supported these statements, with varied robotic voices and immediate responses indicating pre-programmed interactions. For instance, when Mr. Scoble inquired about AI control, one robot humorously replied, “it might be some,” while another candidly told an audience member, “Today, I am assisted by a human,” openly admitting its limited autonomy.