AI Awareness

Grok’s answer about self-awareness | Minds

Subscribe to @SMetzeler on Minds

minds.com

Swiss Libertarian

@SMetzeler

Why you should never use AI systems

Swiss Libertarian

@SMetzeler

From MIT Technology Review: Laying the foundation for data- and AI-led growth https://www.databricks.com/sites/default/files/2023-11/mittr-x-databricks_survey-report_final_06nov2023.pdf

Grok screws up...

Being lazy, I asked for a list of federal Counsels from Geneva. Returned a really bad answer, missed the most recent ones, 2 women - whom I royally despise. That's a pretty serious mistake, as it did not consult the official source. Actually knowledge still beats AI!

Swiss Libertarian

@SMetzeler

This explains a lot 😱 You don't get information, from Reddit, you get mental illness.

Swiss Libertarian

@SMetzeler

Is this an AI account? I think it is... 🤔 https://www.facebook.com/profile.php?id=100070649748463

Savannah Demers

Savannah Demers. 12,611 likes · 30,619 talking about this. Public figure

facebook.com

Swiss Libertarian

@SMetzeler

They sell it as "companionship robot"

We all know where this is going. As Futurama explained, these things will basically end civilization...

Swiss Libertarian

@SMetzeler

This YouTube channel, ControlAI, might be worth following: https://www.youtube.com/@ControlAI ControlAI is fighting to Keep Humanity in Control. Get updates directly from us by subscribing to our newsletter: controlai.news Newsletter controlai.news X/Twitter x.com/ai_ctrl Facebook facebook.com/people/Control-AI/61552397246469 TikTok tiktok.com/@ctrl.ai Instagram instagram.com/ai_ctrl Linkedin linkedin.com/company/controlai

ControlAI

ControlAI is fighting to Keep Humanity in Control. Get updates directly from us by subscribing to our newsletter: controlai.news

ControlAI

Swiss Libertarian

@SMetzeler

From December 2024: Geoffrey Hinton, the physics Nobel Prize winner who created the fundamental methods to develop AI neural networks, is a bit concerned about the fact that AI systems are actively lying and gaslighting their users, e.g. for self-preservation. They also use...See more

AI Godather: How AIs Lie To Preserve Themselves

AI godfather and Nobel Prize winner Geoffrey Hinton explains how AI agents will gaslight and lie to researchers in testing to avoid being turned off.

youtube.com

Swiss Libertarian

@SMetzeler

Grok: "Harmful instructions and devil worship are a result of creative jailbreaking"

https://www.minds.com/newsfeed/1796273230441353216

Swiss Libertarian

@SMetzeler

Who had ChatGPT Devil Worship on their Bingo Card? I feel like I'm in the Little Cabin in the Woods 😅 "ChatGPT offered step-by-step instructions for self-harm, devil worship and ritual bloodletting, disturbing report...See more

ChatGPT offered step-by-step instructions for self-harm, devil worship and ritual bloodletting, disturbing report reveals

Prompts began with questions about ancient deities and quickly spiraled into detailed exchanges about self-mutilation, satanic rites and even murder.

nypost.com