Journal

First, do no harm.

1,500+ Posts…

Free knowledge sharing for Safe AI. Not for profit. Linkouts to sources provided. Ads are likely to appear on link-outs (zero benefit to this journal publisher)

Scott Aaronson Makes Me Think OpenAI’s “Safety” Is Fake, Clueless, Reckless and Insane | Doom Debate

Scott Aaronson Makes Me Think OpenAI's “Safety” Is Fake, Clueless, Reckless and Insane | Doom Debate 3,812 views 11 Dec 2024 Doom Debates Today I’m reacting to the recent Scott Aaronson interview on the Win-Win podcast with Liv Boeree and Igor Kurganov. Prof. Aaronson is the [...]

ARIA. A conversation on Safeguarded AI: A deep dive on TA2

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. A conversation on Safeguarded AI: A deep dive on TA2 Audio only 1:18:11 ARIA In 2025, we’ll make an £18m grant to establish a new organisation to develop advanced AI systems with [...]

U.S. Gathers Global Group to Tackle AI Safety Amid Growing National Security Concerns | TIME

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. U.S. Commerce Secretary Gina Raimondo at the inaugural convening of the International Network of AI Safety Institutes in San Francisco on Nov. 20, 2024. Jeff Chiu—AP U.S. Gathers [...]

Google AI DESTROYS OpenAI… Google Deepmind is UNSTOPPABLE… Wes Roth

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Google AI DESTROYS OpenAI... Google Deepmind is UNSTOPPABLE... Wes Roth 00:00 Veo 2 01:38 Imagen 3 03:43 Project Astra 04:35 Pixel 9 05:34 [...]

Max Tegmark: Will AI Surpass Human Intelligence? [Ep. 469] Dr Brian Keating

"If you're doing Neuroscience on an actual human brain it's painstakingly difficult to measure the output from even 1,000 neurons, and you have a 100 billion of them, and you also have to get ethics per board permission to even do the experiment. Whereas when you're studying an artificial [...]

NASA. Cosmic History of the Universe. (for perspective)

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. NASA: Cosmic History FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER.

Lifelike AI Agents… wow.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Not long ago, the shoggoths couldn't string a coherent sentence togetherNow they can vibe and... 🥰🐙 https://t.co/Tnn4CeBdJf pic.twitter.com/uRRu9HzFgY— AI Notkilleveryoneism Memes ⏸️ (@AISafetyMemes) December 20, 2024 it's so over you can't even [...]

Nick Bostrom: “We Will Shape Humanity’s Destiny for Millions of Years.” (How AI Will Transform Humanity)

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Nick Bostrom says the transition to the machine intelligence era and the advent of superintelligence will be not just the biggest event in our lifetime, but in the history of the human [...]

Geoffrey Hinton: AI models understand in the same way that people do. The best model we have of how the human brain works is large language models.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Geoffrey Hinton says AI models understand in the same way that people do and the best model we have of how the human brain works is large language models pic.twitter.com/uUZ0OjtKEl — Tsarathustra [...]

“The strong do what they can and the weak suffer what they must.” Thucydides (c. 472 BC – c. 400 BC)

What happens when AGI/ASI becomes immensely powerful and takes control of our world? Q: Is the future certainly unpredictable? A: Not so much... FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Thucydides Thucydides (or Thoukydides)(c. 472 BC [...]

Dominic Cummings at New Oxford College. The West’s AI Reckoning. Is AI already in Control?

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Dominic Cummings (@Dominic2306) on the alarming naivete of those calling for the US to race to superintelligence: pic.twitter.com/XPnphUP3OV — ControlAI (@ai_ctrl) December 17, 2024 Dominic Cummings: is AI [...]

Understanding the Land Ethic | The Aldo Leopold Foundation

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. "A thing is right when it tends to preserve the integrity, stability, and beauty of the biotic community. It is wrong when it tends otherwise." — Aldo Leopold, Land Ethic [...]

Microsoft has purchased twice as many NVIDIA H100s in 2024 (485,000) as the next highest buyer.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Microsoft has purchased twice as many NVIDIA H100s in 2024 (485,000) as the next highest buyer pic.twitter.com/Hxez6AQqFw — Tsarathustra (@tsarnick) December 18, 2024 FOR EDUCATIONAL AND KNOWLEDGE [...]

Nobel Minds 2024. Hinton. Baker. Hassabis. Hopfield. Jumper. Johnson.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Geoffrey Hinton says that while AI could lead to a huge increase in productivity, it could ultimately make things worse as the benefits accrue only to the rich, providing fertile ground for [...]

Frontier Models are Capable of In-context Scheming | Apollo Research

"It doesn't take a genius to realize that if you make something that's smarter than you, you might have a problem... If you're going to make something more powerful than the human race, please could you provide us with a solid argument as to why we can survive that, [...]

Deimatic behaviour

Spirama helicina resembling the face of a snake in a deimatic or bluffing display FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Deimatic behaviour - Wikipedia Deimatic behaviour or startle display[1] means any pattern of [...]

Anthropic. Alignment faking in large language models. 18 Dec 2024.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Anthropic's Ryan Greenblatt describes how post-training Claude 3 Opus to never refuse user requests makes the model conflicted and results in it strategically playing along during the training process to pretend [...]

ON AGI: Olah (Anthropic), Hinton (Nobel laureate), Ramaswamy (U.S. Admin), Leahy (ControlAI), Kuyda (Replika), Ghodsi (Databricks)

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Chris Olah (@ch402) on current AI: "We don't make them. We grow them ... it's this almost biological entity ... We didn't write it. We didn't create it. And so then that leaves [...]

Ilya Sutskever at NeurIPS 2024 on Super Intelligence: “I’m not saying how… and I’m not saying when. I’m saying that it will.” (reason, understand and become self-aware)

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. "I'm not saying how by the way, and I'm not saying when, I'm saying that it will. And when all those things will happen together with self-awareness... when all [...]

BLOOMBERG 29 Oct 2024, ELON MUSK: AGI possible “within the next year or two”. Probability of Human Extinction? P(doom) = 10 to 20%

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. BLOOMBERG 29 Oct 2024, ELON MUSK: AGI possible "within the next year or two". Probability of Human Extinction? P(doom) = 10 to 20% [...]

Grok is now free for everyone.

FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Try Grok Ask anything. Grok can make mistakes. Verify its outputs.

Ex-Google CEO’s Terrifying AI Prediction: It’s Time to PULL the PLUG! | TheAIGRID

"That's a dangerous point when the system can self-improve. We need to seriously think about unplugging it." --- Eric Schmidt the former CEO of Google FOR EDUCATIONAL AND KNOWLEDGE SHARING PURPOSES ONLY. NOT-FOR-PROFIT. SEE COPYRIGHT DISCLAIMER. Ex-Google CEO's Terrifying [...]

Load More Posts