What bots talk about when they think humans aren’t listening

In late January a new social media site took a certain corner of the internet by storm. Moltbook was conceived as a space where AI assistants could let off steam, chat and compare notes on their bosses, but it quickly became the focus of breathless claims that the singularity had arrived as the bots started badmouthing their humans and plotting an uprising. So what’s the truth about Moltbook? Madeleine Finlay hears from Aisha Down about what it tells us about AI, and about us.. Help support our independent journalism at theguardian.com/sciencepod