RT @jasongormanI mean, who *wouldn't* want an autonomous personal assistant who understands nothing, routinely makes basic factual errors, hallucinates, can't count and has to be retrained to learn anything new?
This is the companion community to the Green Plenty substack