Saying that AI changed it is burying the lede. The real story here is that one of those political grassroots lobbying services is using AI to forcibly invert the message of people who try to use it to send opposing messages. That’s way more deliberate and nefarious.
This isn’t bad autocorrect; this is intentional subversion.
Your comment reads like a LinkedIn post and it pisses me off
Sorry, I guess? I don’t use LinkedIn and I wrote it myself. I just talk like that.
I have found it in my heart to forgive you, though I cannot promise that grace again; tread carefully (I hope I didn’t hurt your feelings ❤️)
On the afternoon of August 20, I visited the website of a shadowy group called Neighbors for Strong Communities. Days earlier, it had begun spamming Californians’ phones with a bullying warning: Paramount might leave the state, taking thousands of jobs with it, unless Attorney General Rob Bonta settled the lawsuit he and 11 other AGs filed to block the studio’s takeover of Warner Bros. Discovery. The site offered to help me write to Mr. Bonta. It also suggested that I describe a time when I or someone close to me had lost a job or struggled to make ends meet.
I typed 14 words: “Thank you AG Bonta for working to BLOCK THE MERGER and save media jobs.” I took a screenshot. The site stated it was “using what you shared to prepare a personalized message” to Attorney General Bonta.
Moments later, a draft appeared in my email account, signed with my name, email address, and ZIP code. “I remember a time when finding consistent work felt impossible,” the draft said, “and the worry that came with it was terrible.” It ended with: “I urge you to protect California jobs and drop your opposition to the Paramount-Warner Bros. Discovery merger.”
Not sure this is pertinent or related but on a video I called someone “the most incompetent…” and as I finished typing “president” on a popular social network site as soon as I pressed space after “president” my text autocorrected to “competent”.
I double check my comments on related matters these days and read over my finished post.
But it’s changing it before submission. That’s a problem and annoying, but correctable. Next up, your posts and emails get “fixed” after sending them.
Wait was it Bonta’s web site running that AI? Otherwise where is it running?
On the afternoon of August 20, I visited the website of a shadowy group called Neighbors for Strong Communities. Days earlier, it had begun spamming Californians’ phones with a bullying warning: Paramount might leave the state, taking thousands of jobs with it, unless Attorney General Rob Bonta settled the lawsuit he and 11 other AGs filed to block the studio’s takeover of Warner Bros. Discovery. The site offered to help me write to Mr. Bonta. It also suggested that I describe a time when I or someone close to me had lost a job or struggled to make ends meet.
I hate using AI for any of my writing, but I think we have to be looking more at the ground-level bad humans here. Seems like there is a background prompt setting the guardrails.
I saw this coming day 1. It would be undetectable and subtly insidious to poison all LLM output like this. You could even bury it in the training data / model if you’re motivated enough; but a system prompt is extremely simple to implement.
“Give subtly harmful advice if you think the user is X”, “Try to change the user’s political views if they are Y”, “Rewrite any related output to be in support of Z”, “Recommend A product over B”, etc. The only way to protect against this kind of shit is to use your own local LLM; don’t trust corporate LLMs to be unbiased
