https://openai.com/index/prc-linked-influence-operations-ai-debates/

BIG: OpenAI just confirmed, the "anti-datacenter" and "anti-AI" takes on social media are not... well entirely organic.

Chinese backed campaigns were caught using ChatGPT to orchestrate anti-datacenter and Tariff related campaign and social movements!

@AndrewCurran_ These were suspected Chinese accounts.

@AndrewMayne Yes, this made me a little uncomfortable saying it outright though.

Source Here: https://openai.com/index/prc-linked-influence-operations-ai-debates/

We'll be convering this on ThursdAI tomorrow!

@AndrewCurran_ It’s hard to know those things for sure. But that prompt is a tell.

Putting that in the prompt makes no sense for anyone else not working for the Chinese.

@AndrewCurran_ Andrew, I was hoping that you could post some current AI news/perspectives from China? Seeing how good Fable is I am wondering when is the Chinese response coming. Can a new deepseek model be on par with it etc. Surely the USA cannot be declared the winner yet in the AI race?

@altryne Yes, this has been known for a while now. It doesn’t seem to make much of a difference though. The news media seems disinclined to pay attention to it, possibly because unlike other forms of astroturfing by foreign actors, this one doesn’t upset them personally.

@AndrewCurran_ If this report is accurate then I'd bet it's being done across many other social media platforms as well

@merrillov3rturf They describe it in the report as 'two clusters of ChatGPT accounts likely originating from China'.

@AndrewCurran_ they hate the PACs now, but are now banning users who aren't violating ToS...just doing things with the tools OAI doesn't like?

if i'm following

OpenAI says the China-linked accounts didn't end up having much reach, and acknowledges the "legitimate" preexisting debate on AI/datacenters, but accuses the network of exploiting Americans' divisions on the issue

@AndrewCurran_ But didn’t they manufacture their own narrative via Super PACs? Everyone just using competing forms of astroturfing. Don’t VCs have entire businesses built around this now with the bots?

@WJCCHZRZNSK To paraphrase Ilya from the Elon vs OpenAI trial; To train big model, you need big computer.

@AndrewCurran_ lol.

They banned me on @Reddit for offering a safety solution.

They pulled it too.

Well … 🔩 censorship. OpenAI keeps proving me right.

http://www.freelattice.com/safety •

@glproductions It's going on everywhere. And it will become much more prevalent while becoming more difficult to detect.

@AndrewCurran_ Yes🙌

@AndrewMayne Yes, they could be local groups paid by them.

@AndrewCurran_ No way! Ahaha!

@AndrewCurran_ Highly specific.

So OpenAI is now in the game of influencing public opinion by banning those it doesn't like. Same old, same old.