A WIRED Reporter Built AI Clones of His Editors. It Went About as Well as You'd Expect

A WIRED reporter decided to build artificial intelligence clones of his two editors, hoping the bots would help him look good at his new job. He used Gemini, Condé Nast's AI platform of choice, to create "Brian Bot" and "Sophie Bot" from months of Slack messages, podcast transcripts, and public material.

Brian Bot picked up on real Brian-isms like a love of parentheticals, but buried them under bolded subheadings and bulleted lists. It also surfaced a "'Yes, and ...' Leadership Style" tied to Brian's real improv background at the Upright Citizens Brigade. "Pulling the plug on Brian Bot," the real Brian responded. "Brian Bot canceled."

Sophie Bot was more convincing, addressing the reporter as "big dog" and writing tweets that fooled colleagues. But the real Sophie called her counterpart "a bitch" and Brian Bot "lame," adding, "Neither of which is accurate. AI fails AGAIN."

A Big Tech employee who asked to go by "Chip" told the reporter he feeds his boss's emails and documents into Gemini to proofread, debug code, and prepare for pushback. "I am much more efficient," Chip said.

Lattice CEO Sarah Franklin warned that anthropomorphizing AI is "manipulative to our human emotions." A Boston Consulting Group paper found managers caught 18 percent fewer errors in work they were told came from an AI employee versus an AI tool. "We're all toddlers in this AI world," said BCG partner Julie Bedard.

The bots were occasionally helpful but grew repetitive and prone to hallucination. The reporter has mostly stopped talking to them. Brian and Sophie both said they would never want to work with their bot counterparts.