using FOIA requests to get LLM chat history like this is a journalistic angle I haven’t seen before, it’ll be interesting to see it get picked up elsewhere.

I imagine AI boosters will defend this as being totally fine and normal and a good use of LLMs…but if it was, I’d expect more compliance with a policy about saying “this was LLM-generated”. if the tool is so useful, why not brag about how much you’re using it?

In July 2024, Everett’s IT department sent guidance to city employees saying all AI-generated material “released to external audiences for the purpose of public policy decision making should be clearly labeled as having been produced in whole or in part by AI.”

“I think people would want to know” when AI is used in government, Franklin said.

But staff haven’t followed the guidance consistently. Records show several examples of unlabeled AI-generated content created after the guidelines were released.

In one instance this spring, Franklin sent a letter to U.S. Rep. Rick Larsen urging him to co-sponsor the DRONE Act of 2025, which would have made it easier for cities to buy drones for law enforcement. The mayor’s letter was entirely generated by AI, based on a three-sentence prompt from a member of her administration. It makes no mention of this.

possibly including getting a “personalized” response from a letter you send elected officials:

When a senior citizen in Bellingham emailed about not being able to afford utilities, a city official pasted his email into ChatGPT and asked for “a sympathetic response.”

“Thank you for taking the time to share your concerns so clearly and thoughtfully,” the AI-generated reply read. “I hear your frustration.”