How Do You Change a Chatbot’s Mind? When I set out to improve my tainted reputation with chatbots, I discovered a new world of A.I. manipulation. (www.nytimes.com)
from silence7@slrpnk.net to technology@lemmy.world on 30 Aug 2024 17:46
https://slrpnk.net/post/12866753

#technology

threaded - newest

conciselyverbose@sh.itjust.works on 30 Aug 2024 18:14 next collapse

My theory about what happened next — which is supported by conversations I’ve had with researchers in artificial intelligence, some of whom worked on Bing — is that many of the stories about my experience with Sydney were scraped from the web and fed into other A.I. systems.

These systems, then, learned to associate my name with the demise of a prominent chatbot. In other words, they saw me as a threat.

🤦‍♀️

Antmz22@lemm.ee on 30 Aug 2024 21:19 next collapse

Most trustworthy and accurate New York times article.

LostXOR@fedia.io on 30 Aug 2024 22:50 collapse

I'm tired of people ascribing any sort of intelligence to AI. It's not thinking, it's not seeing you as a threat, it's just predicting a probable response based on its training data.

Deceptichum@quokk.au on 30 Aug 2024 19:32 next collapse

This guy is a moron.

If the bots are saying they hate him and that he sucks, it’s because that’s what the general consensus was from all the data they scrapped not because the bot is scared of him as an AI killer.

silence7@slrpnk.net on 30 Aug 2024 19:53 collapse

The bots are not reliable summarizes like that. They often can’t tell the difference between the author and the subject of a piece of writing.

BelatedPeacock@lemmy.world on 30 Aug 2024 23:35 next collapse

My theory about what happened next — which is supported by conversations I’ve had with researchers in artificial intelligence, some of whom worked on Bing — is that many of the stories about my experience with Sydney were scraped from the web and fed into other A.I. systems.

These systems, then, learned to associate my name with the demise of a prominent chatbot. In other words, they saw me as a threat.

LLMs predict text, they don’t have feelings or awareness. Even if a researcher did say that I call to attention the Google chatbot programmer who thought an LLM became sentient because it said so when generating text.

Guys, my paper is sentient, it says so.

If the AI says he’s disonhest and sensational that’s because enough people on the internet have said so that the AI considers it to be true.

silence7@slrpnk.net on 30 Aug 2024 23:40 collapse

It doesn’t take people on the internet saying it though; just an association with people saying something and the name, which happens to people who write news articles about something.

AwesomeLowlander@sh.itjust.works on 31 Aug 2024 00:23 collapse

Prompt Google’s Gemini for its opinion of me, and it may respond, as it did one recent day, that my “focus on sensationalism can sometimes overshadow deeper analysis.”

Based on this article, it turns out the chatbots do get things right sometimes. The rest of his article made it pretty clear he’s aware that they’re not intelligent, but he just couldn’t resist the sensational opening of ‘AIs hate me!’