I’ve been thinking about this for a few days now and I think there is a big missed opportunity here. The holodeck prototype showed more of less the same errors holodecks still have 100 years later. They could have shown other faults and use that for a comment on the current AI development.