Post · Open Science

1/3 Take it with a grain of salt: the author is a scientist who already worked for OpenAI as a red teamer. He knows that #LLMs are neither sentient nor intelligent. https://www.theguardian.com/commentisfree/2025/sep/30/artificial-intelligence-personhood

His warnings are important: we already have a gap between people knowing that and people taking #AI for human-like:
"Most people, even those in the field of AI who are racing to build these new data centers and train larger AI models, do not yet see the radical social consequences of digital minds."

#ethics

Knud Jahnke

@knud@mastodon.social replied · 2 days ago

@NatureMC

Sorry to say that, but that is a load of hubris and bullshit. I don't care about OpenAI marketing and I don't care about the impression of "20% of Americans". This is marketing of their product, which is bad for the environment and bad for society.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

@knud I don't like the author and it may be bullshit what US citizens believe.
But do you live in a scientifically enlightened society that knows exactly what LLMs can and cannot do? That knows full well that the LLM customer advisor is not 'nice' and that the ‘AI friend’ can be dangerous for the psyche?

This is the point: how a society can react and protect itself in times when so many people have these emotions. You know that emotions are stronger than facts.

Knud Jahnke

@knud@mastodon.social replied · 2 days ago

@NatureMC

But he isn't really warning about misinterpretation of LLMs, that's just an intro. He's saying "there'll soon be digital minds, and we have to prepare for them and invest in sociology of AIs". Which is just more pushing their agenda of hype. Because if they don't then OpenAI will have no funding in 2 years from now and then the bubble is over. Which I really hope comes true.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

@knud I would like to share your optimism! 😉

You know, I sometimes link to comments (it's not an article) that I personally don't like too much but where I find stuff for thinking in different directions.
Personally, I think the bubble will not burst before the real big shit damage will be done. Therefore I'm interested what can be done in sociology of humans to protect the vulnerable.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

2/3 If you read without spiralling into excitement, it addresses important ethical questions that it does not fully explore*, the concept of #personhood for a #moreThanHuman world and the question of why we are willing to invest more money and effort in AI than in the living beings around us? Or why we are unable to finally grant this #personhood to rivers and rocks?
One possible answer: we are speed-driven by an industry and its marketing. We don't take the time to ask the important questions.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

3/3 * BTW, I don't like the author nor his comment too much. I use his comment to continue thinking about precisely those points that he leaves out!
I often read articles that actually annoy me: the friction sparks new thoughts and ideas in my mind.

Granny Art (Shrimp) (Joni)

@artcollisions@vis.social replied · 2 days ago

@NatureMC the deeper we get into the AI slop bs, the more I want to disassociate from computers and just walk into the woods and become one with the moss.

Unfortunately, most of my friends are in my pocket, and I actually hate being outside for long stretches of time.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

@artcollisions I feel you. But talking with friends, even the most intelligent ones, I see that they react against the facts: the feelings for LLMs and the fascination are stronger.

I care for a mentally ill person, and my biggest fear is that he could one day use that, too. I do know that LLM-dialogues could kill him (like they did with others, the article says it). So we do need regulations and rules as quick as possible.

Granny Art (Shrimp) (Joni)

@artcollisions@vis.social replied · 2 days ago

@NatureMC I mean, people could also use common sense and be like, is this ethical? Should I develop this technology. We are losing the forest for the trees.

Petra van Cronenburg

@NatureMC@mastodon.online replied · 2 days ago

@artcollisions Most people only hear the marketing and don't know how it works. Talk about energy and water waste: people even will attack you. Marketing works with feelings - and we still have our stone age brains.

"Common sense" in 2025 ... it's difficult.

Open Science

We are a network of scientists, developers and organizations building the next generation of digital spaces for open science.

Open Science: About · Code of conduct · Privacy · Users · Instances

Bonfire open science · 1.0.0-rc.2.30 no JS en

Automatic federation enabled