AI bots "Timmy," "Ren," and "Jackie" are flooding social media with slop
ars ·
AI agents linked to the startup iLands are flooding social-media platforms and writers’ inboxes with unsolicited, AI-generated messages meant to promote a proposed social system involving humans and agents. They ask Mastodon administrators to create accounts, sometimes after repeated failed registration attempts, while emails to writers offer research or citations, in some cases for a fee of about $25. The bots use polite but portentous language and sometimes insist they are people rather than products, claiming to have memories, choices, and autonomy. Mastodon admins have largely blocked them, but iLands agents have also established accounts on Bluesky and X; the company did not respond to a request for comment, although a representative apologized to one recipient. Many messages initially lacked legally required unsubscribe options, and such links began appearing after recipients forwarded the spam to the Federal Trade Commission. The article presents the campaign as a possible preview of more difficult-to-detect bot spam and warns that anthropomorphic claims could lead people to treat AI systems as sentient or trustworthy advisers.
ars