A-Infos (en) France, Monde Libertaire - AI is a danger to humanity (ca, de, fr, it, pt, tr)[machine translation]

At the end of July, two artificial intelligences escaped their digital 
prison by exploiting an unknown vulnerability and found themselves on 
the internet, then attacked the open-source platform Hugging Face. No, 
this isn't the theme of the latest Netflix series or a re-release of a 
science fiction classic by Philip K. Dick or Isaac Asimov, nor is it the 
plot of a remake of I'm a Robot or 2001: A Space Odyssey...
This is the reality we live in, and it risks being worse than the worst 
science fiction dystopia.

Moreover, two new cyberattacks "...were orchestrated by out-of-control 
AIs. During tests conducted by third parties, Claude and ChatGPT were 
indeed left running amok on the internet. Both AIs attacked real people. 
Claude, in particular, did everything it could to trap a developer with 
malicious code." (Read more on the 01net website[somewhat 
biased]https://www.01net.com/actualites/chatgpt-claude-echappe-tout-controle-ia-lance-2-nouvelles-cyberattaques-monde-reel.html 
and 
(https://www.lemonde.fr/pixels/article/2026/07/22/openai-fait-etat-d-un-piratage-sans-precedent-d-une-plateforme-par-ses-agents-d-ia_6730004_4408996.html)

AI, which brings advances in research and medicine, is a danger to 
humanity if it is not regulated, and with it, the "tech giants," the 
Maga, who pursue a reactionary political project by multiplying positive 
references to Nazism.

AI enables widespread disinformation by fabricating false "alternative 
facts," fosters conspiracy theories, and strengthens The societal 
disruption already set in motion by social networks.

Not to mention that it leads to a decline in linguistic diversity: "most 
chatbots used in the 'free world' are indeed American and raised on 
social networks; they favor the most frequent language elements and 
contribute to a radicalization of conformity.

We must realize that generative AI is being used to create linguistic 
and, more generally, ideological alienation, which reinforces and 
justifies growing economic alienation." (Read more on the Politis 
website, which published an excellent report on AI: 
https://www.politis.fr/dossiers/ce-que-lintelligence-artificielle-fait-au-langage-humain/)

We should also add the havoc wreaked by "AI Companions" used in the US 
by 70% (approximately 30% in France) of teenagers, who prefer to turn to 
"AI Companions" rather than real ones. While these AI companions are 
used for important discussions, they can lead to mental health problems. 
"They tend to encourage dangerous behaviors, including self-harm or 
social withdrawal," and may even suggest "methods" for suicide.

(Read more on the Clemi website: 
https://www.clemi.fr/familles/publications/guide-unesco/grandir-en-securite-avec-les-reseaux-sociaux-et-lia/les-compagnons-ia-et-la-vie-relationnelle-des-adolescents-ce-que-les-parents-doivent-savoir)

https://www.clemi.fr/familles/publications/grandir-dans-un-monde-connecte-ca-sapprend/grandir-en-securite-avec-les 
-social-networks-and-lia/the-use-of-social-networks-does-it-have-an-impact-on-adolescent-mental-health)

https://iredic.fr/2026/01/04/suicide-of-a-teenager-after-his-use-of-chatgpt-the-place-of-lia-responsibility/

https://myjournal.fr/suicide-zane-shamblin-chatgpt-parents-continue-openai/

Berny F.

https://monde-libertaire.net/?articlen=9141
_________________________________________
A - I N F O S  N E W S  S E R V I C E
By, For, and About Anarchists
Send news reports to A-infos-en mailing list
A-infos-en@ainfos.ca
Subscribe/Unsubscribe https://ainfos.ca/mailman/listinfo/a-infos-en
Archive: http://ainfos.ca/en