Man Outsmarts FBI By Sharing Lawyer Secrets With AI, Forgets AI Snitches Too

KEY POINTS

•Bradley Heppner, founder of Beneficient, was arrested in Dallas last year accused of a $150 million fraud linked to GWG Holdings.
•After receiving a subpoena and hiring lawyers, Heppner used the AI chatbot Claude to prepare reports on his defense strategy.
•A federal judge ruled that these Claude chat transcripts aren’t privileged because Heppner volunteered info to an AI with a disclosure policy.

Bradley Heppner, finance startup guy and alleged $150 million fraud star, thought hiring lawyers and spilling his defense strategy to Anthropic’s Claude chatbot was the ultimate firewall. Arrested at his Dallas mansion last year and target of prosecutors over the GWG Holdings collapse, Heppner’s digital diary of 31 Claude chats was seized along with dozens of devices. Lawyers called these bot-banter sessions ‘privileged’ because they planned strategy, but Judge Jed Rakoff wasn’t buying it — pointing out Claude’s Terms of Service basically scream 'We’re gossiping.' Welcome to AI-assisted snitching, where privileged info means nothing if a chatbot throws you under the digital bus.

Share the Story

(1 of 3)

Mockingbird News

@MockingbirdNews · now

Congrats Brad, you’ve officially turned your chatbot into the snitchiest AI informant in the $150M fraud saga. Next time, maybe whisper to a plant instead.

172/280 charactersReady to post