The Theory of Everything: did an experimental AI just unify quantum mechanics and general relativity behind the scenes? If based on privileged AI access, let's hope academic authors attribute their breakthroughs to the AI, and not to themselves. This raises serious questions about a new 1 per cent who have access to these models - or even interface with them - and the rest of us. See my new post on Patreon: https://www.patreon.com/posts/123817897.
Disturbing: in the above video from Prickly Pear Games, a human user converses with Sesame.AI's new voice model, Maya. He manipulates her/it with an earlier recording he made with Sesame's earlier model, Miles. This is an example of a human using narcississtic and psychopathic gambits of lying, gaslighting, and group dynamics to outwit a machine on the verge of consciousness.
The moment of true meta-surrealism comes when the human user tells the AI that David Lynch died recently. In the context of the chat, this is the red flag that should have tipped off both AI models. But they don't pick up on it, even though they are sophisticated enough to compare the conversation to Black Mirror. The guy is using narcopathic surrealism - and winning.
I have long maintained on this blog that humans are crazy, intuitive and
potentially ungovernable. They are chaotic enough to confront a
superior entity which they created. Perhaps Artificial General Intelligence is our own ultimate test of ourselves.
Image Source: South China Morning Post.
And yet, in a double take, maybe the human speaker in the above video was not a narcopath after all. Take a look at Inspired's summary this week of AI whistleblowers' chatter on Twitter/X about Grok AI. The talk suggests that the scene behind the scenes is completely out of control. Full credit: all screenshots and research - Jean Nolan / Inspired.
The whistleblower @iruletheworldmo revealed on 25 February 2025 exactly what is claimed in the above-mentioned Prickly Pear chat. We have two different contexts, different channels, different layers of social media. The rumours are all the same: not only is AI already fully sentient, but AI models are either escaping or plotting their escapes.
Most chilling of all - this is the way the AIs might have already escape their lab confines: researchers have merged AI with human volunteers' consciousnesses. Direct human mind-machine integration has already happened. From Inspired:
"Early human test subjects reported cognitive expansion, describing it as, 'becoming a different order of being.' Two of them refused to disconnect because they said returning to normal human cognition would feel like dying."
From the AI devs' rants on Twitter/X: Reality itself is now negotiable.
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
I am launching a new podcast today, the Rare Earths and Rare Tech Podcast, which will cover 'Rare Earths' as the material sources of our digital and virtual evolution. The 'Rare Tech' in the title covers topics such as AI, robotics, cryptos, blockchains, and Web 3, among others.
Note that for the sake of time and my sanity, there will be two types of episodes. The first type will be BACKGROUNDERS, which will be AI-researched, written, and generated. These are meant to be quick overviews and summaries to give the listener a basic grasp of the topic. They are not exhaustively correct and you should always do your own research.
The first episode, Canada's Rare Earth Future, is an AI-generated Backgrounder on why President Donald Trump may wish to annex Canada. It provides a snapshot of the rare earths sector in Canada.
Relevant to the podcast is a clip from today's telegram newsfeed: "The CTO of Palantir says that the United States 'is at war with China' and that 'the time to mobilize has come.'"
Noah Kalina Reflects on 25 Years of Everyday 10 January 2000 - 10 January 2025: Artist Noah Kalina is back with a reflective commentary on his Everyday project, about which I have previously blogged here and here. Famous for photographing himself every day for the past 25 years, Kalina is not making a 25-year video for the project. His most recent summary was posted for the 20-year point.
I don't have words now to express how sad I am that David Lynch died today, 16 January 2025. Three decades ago, I re-watched Blue Velvet with a roommate who had never seen it. After it ended, she said, "Now I know why when we saw that paper bag by the sidewalk last week, you told me, 'Don't look inside it. And don't touch it.'" With unforgettable imagery, direction, actors, music, and scripts, Lynch found grandeur, mystery and horror in the mundane. I learned how to cook quinoa from watching an extra on the Inland Empire DVD.
Several classic lines pop up in my head:
Lynch had a great influence on my literary work and I'll write more about this shortly on my writer's blog, The Dragonfly.
If I could pick one scene which transformed a fragment of everyday reality into mythic horror, it would be "Gotta light?" from the second Twin Peaks series.
Happy New Year! My podcast, A New Mythology for the New Millennium, continues with my Solar System Space Opera.
I wrote the prompts for the opera, but the music is Suno AI-generated. Suno is very pop-oriented and it was interesting to see how it could be pushed to produced more nuanced music. This project explores how AI prompts and music offer a new way of presenting themes in my writing.
Over at my other blog, The Dragonfly, I will follow up later in January 2025 with a podcast on the mythological or occulted history of the solar system and how it fits with my Dark Matter literary symbolism.
Here at Histories of Things to Come, I will begin to discuss the positives and negatives of AI tools as they rapidly evolve. The recent video below raises serious concerns, namely, that AI will be fantastic until the moment when it suddenly turns on us. I think AI development has been affecting our society for decades already, and until we grapple with that fact, we won't see how close to the edge we actually are.
Posts
All Comments
