Typhus's Tavistock Testing Center

I expanded on my previous posts about AI for this article on RadicalDose.com https://radicaldose.com/ai-is-no-big-deal-or-laundry-buddy-isnt-coming-to-kill-you/

THE PARANORMIES PRESENT: RAGTIME REVELATION OF THE METHOD WITH FOUNDRING S09 EP31 🔗 Website Link |🔗RSS Link |🔗Pilled Livestream Page SEASON 9 CONTINUES!!! Johnny and Reinhardt are joined by the (unbeknownst to him) writer and performer of the official Paranormies theme song “Everything is Fake and Gay”, the one and only Foundring! The conversation winds around a myriad of subjects from music nerdiness, vibrations, occult music and musicians, conspiracy theories, Sandy Hook, the theory of the occult, Pizzagate, 9/11, personal conspiracy journeys, schizo fellowship, and much more! Get /comfy/ and let the ragtime roll! @TheParanormiesPresent @ApeMtn38

Bill Walton, who boasted on air about getting his second Moderna dose, dies of cancer.

3/3 on this graph, here’s how I see 3 different scenarios for how AI could progress over time, with the optimistic openAI stockholder shill scenario represented in yellow as exponential progress, and a linear progression represented in red. It’s important to note that literally zero technologies have ever progressed over time in a linear fashion. The scenario I think we’ll experience instead is outlined in green, very quick progress that seems like it’s on an exponential curve in green, which just as quickly flattens out where as time progresses you end up seeing very little further progress.

3/3 on this graph, here’s how I see 3 different scenarios for how AI could progress over time, with the optimistic openAI stockholder shill scenario represented in yellow as exponential progress, and a linear progression represented in red. It’s important to note that literally zero technologies have ever progressed over time in a linear fashion. The scenario I think we’ll experience instead is outlined in green, very quick progress that seems like it’s on an exponential curve in green, which just as quickly flattens out where as time progresses you end up seeing very little further progress.

2/3 This is an example of an Alan Moore script. One panel. For one issue, Moore might turn in 100 pages+ of this. Moore has worked with dozens of artists in this fashion. To get to an AI model that could just work with Alan Moore to draw a comic how he likes it, you’d probably need more silicon than exists on the planet.

🧵1/3 One of my most contrarian takes in conspiracy circles is that I don’t actually believe “AI” is any real threat at all to basically anything, that’s there’s no secret versions of AI models lurking behind the scenes or buried in a CIA datacenter and that what we see is just what they let the plebs have, etc. There’s several companies that exist that are pursuing development of “AGI” or “Artificial General Intelligence” as part of their mission statement. OpenAI is probably the biggest one. I think, on a technical level that you can actually prove with evidence and experimentation, that this is an impossibility and that their stated belief otherwise is a type of optimistic self delusion, but more probably, a cynical effort to milk investor cash and exit scam with it. I expect AI development to run into a singularity of complexity. That is, while the first chunk of the work towards this goal might come quickly, at a seemingly exponential rate, the progress quickly levels out and reaches a point where the complexity of further progress starts to become so incredibly complex that the difficulty of further progress starts becoming exponentially more difficult, until you reach a point where that difficulty approaches infinity. An example of this that you can test experimentally is the problem of imprecise vs precise work. AI is quite good at doing work where there is a wide tolerance for imprecision. If you ask an AI model, draw me a picture of a car dashboard, it can do this perfectly well. Ask it, draw me a picture of a 1993 BMW M3 dashboard with 85mph on the speedometer, 5000rpm on the tachometer, and 40% fuel in the tank and it quickly fucks up and can’t do any of that, and this is only asking for slightly more precision. The datasets that you need to train AI models on to be able to produce precise work becomes orders of magnitude more large with every additional step of precision you introduce. At a certain point, they become impossibly large. Take this example of the car dashboard, now imagine it’s one small detail in a drawing you want to render where there’s 20 cars on a street, all different makes and models, unique details about each, their drivers, the trees along the road, the buildings, etc. Then extrapolate this further and imagine you’re trying to render a Grand Theft Auto style game with 11,000 streets. New York City has 32,000 streets for reference. Alan Moore, the comic writer, is notorious for producing exceptionally precise scripts. He’ll use an entire page of text or more just for one panel of a page, dictating precisely what he wants to see in the panel to an artist. This has pissed his editors off a few times, but he’s been able to work for decades in this manner. Feed one page of an Alan Moore script to an image generation model and it will completely shit itself.

Lmfao