The following article was first published on Shipman & Goodwin attorney Dan Schwartz’ Connecticut Employment Law Blog. It is reposted here with permission.
If you’ve been following this blog, you know I’ve been writing about the intersection of generative AI and employment litigation for a while now.
I’ve talked about updating litigation hold policies to account for GenAI data, and I’ve urged employers to start requesting plaintiffs’ AI conversation histories in discovery.
Well, a ruling this past week from the Southern District of New York just added an exclamation point to all of it.
On Feb. 10, 2026, Judge Jed Rakoff ruled in U.S. v. Heppner that documents a defendant created using Anthropic’s Claude AI tool and later shared with his attorneys are not protected by attorney-client privilege or the work product doctrine.
The defendant, a financial services executive accused of fraud, had used Claude to prepare approximately 31 documents related to his legal situation and then sent them to his defense counsel at Quinn Emanuel.
When the government seized his devices and found the AI-generated materials, the defense tried to shield them as privileged.
Judge Rakoff wasn’t having it. “I’m not seeing remotely any basis for any claim of attorney-client privilege,” he said from the bench.
The reasoning is straightforward and grounded in well-established privilege principles: An AI tool is not an attorney. It holds no law license and owes no professional duties. The tool’s...
A disinformation campaign which has been linked to pro-Russian actors is imitating reputable media outlets to push fake reports about the Middle East war, as part of a campaign to discredit the W...