Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

IbogaProject

(5,906 posts)
14. The models are learning from him
Sun Mar 29, 2026, 08:40 AM
Sunday

So his prompts are being injested by the models. If he is auditing output with another model it is learning too. Anerica's prosperity is near collapse. Our treasury has admitted the federal government is now insolvent. Taco insulted Japan during a crucial meeting where she came to get concessions to continue to only buy oil in Dollar denomated transactions. They have now decided to by Yuan based contracts, which will increase the chance that Iran will let those ships past. Japan is by far the largest holder of American Treasuries. If and when interest rates increase those securties will loose value. There will be a run on our debt and it will set off a Nineteenth century style depression. Rubio already said this not-war's outcome will be Iran tolling for passage across the straight. Our AI is cost constrained, China has been releasing stuff opensource I see huge disruptions coming. Throw in a climate break with runaway methane releases and then major earthquakes and one or more mega volcanos and things could move into catastrophe. I think this AI bubble is just a symprom of a system bluffing its way through recent financial quarters.

Recommendations

1 members have recommended this reply (displayed in chronological order):

What if they... 2naSalit Saturday #1
It's like the fuckin tulip mania of the 1630's struggle4progress Saturday #2
Bingo. multigraincracker Sunday #13
AI can only copy what has been done edhopper Saturday #3
The layoffs ARE about AI. These companies have all pissed away hundreds of millions of dollars Bluetus Sunday #6
Just to illustrate that point about curated LLMs Bluetus Sunday #16
Deep expose. cachukis Sunday #28
Very well explained. Thanks. cachukis Sunday #27
I think that if an individual has some knowledge or expertise and asks the right questions that it can lead CentralMass Sunday #9
The difference is edhopper Sunday #15
I understand your point, but i think that people who think that they are safe from it are being overly optomistic. CentralMass Sunday #18
I don't think anyone is safe edhopper Sunday #20
Are psychologists relatively safe? Dave says Sunday #25
+! struggle4progress Sunday #12
It's a lot more nuanced than that. tinrobot Sunday #17
Yes edhopper Sunday #21
Codex is beyond that. It has special tools built into it to automate coding that go beyond just being trained... FascismIsDeath Sunday #22
I get that edhopper Sunday #24
When circumstances are forcing you to lay off people... hunter Saturday #4
A year from now: they'll be hiring people back to debug AI slop code. JHB Saturday #5
I'm not sure about that. From what I can tell, it is getting better and better. With a knowledgeable programmer working CentralMass Sunday #10
It seems to be a bit soft, but I suspect it's more gristle and chewy. haele Sunday #7
As an immersed peruser of history, I like your study cachukis Sunday #29
Meanwhile, companies are hiring human programmers to fix all the shit AI does sakabatou Sunday #8
An AI response to a related question. CentralMass Sunday #19
Yeah, who's going to organize your fantastic new army of agentic AI? Gonna do it all yourself, Sam? 0rganism Sunday #11
The models are learning from him IbogaProject Sunday #14
I agree, that is a concern, one only partially addressed by exclusively running locally. 0rganism Sunday #31
I see your point. But the scenario id that the compamy employs a small team of very skilled coders abd ir engineers who CentralMass Sunday #23
It's definitely an awful situation, but there's an up-side too 0rganism Sunday #30
Five to one, baby, one in five... OC375 Sunday #26
Latest Discussions»General Discussion»Futureism- Sam Altman Tha...»Reply #14