Lolwut: “Was the Iran War Caused by AI Psychosis?
-
Lolwut: “Was the Iran War Caused by AI Psychosis?
AI simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.”
Was the Iran War Caused by AI Psychosis? | House of Saud
AI sycophancy, RLHF bias, and Ender's Foundry simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.
House of Saud (houseofsaud.com)
RE: https://neuromatch.social/@jonny/116325053467318924
Did you catch the news earlier that Claude was being used to provide targeting information?
I can see why Anthropic was having a shit fit over legality, but the totalitarian guy running the White House fretting over laws as woke.
-
Lolwut: “Was the Iran War Caused by AI Psychosis?
AI simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.”
Was the Iran War Caused by AI Psychosis? | House of Saud
AI sycophancy, RLHF bias, and Ender's Foundry simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.
House of Saud (houseofsaud.com)
The trouble with a decapitation strike is if your enemy has broken itself down into multiple cells and has commands to 'keep following most recent commands until a next leadership issues new orders'.
-
RE: https://neuromatch.social/@jonny/116325053467318924
Did you catch the news earlier that Claude was being used to provide targeting information?
I can see why Anthropic was having a shit fit over legality, but the totalitarian guy running the White House fretting over laws as woke.
@GhostOnTheHalfShell yeah I saw that
-
@kcarruthers On the one hand, this explains a lot of things. Like how SECDEF didn’t understand that bombing the capital of a country wouldn’t win them friends, and that Iran—who famously sent children to the front lines in the Iran Iraq war—wouldn’t fold easily, and would be willing to suffer the closure of the strait of Hormuz.
On the other hand, holy shit! What the actual torment nexus fuck?
Thirdly, the US is now considered an unreliable ally (although ask the Kurds…)@ShiitakeToast IKR YMMV with USA
-
@Rataunderground @kcarruthers
Not all false beliefs reach the level of psychosis. Believing someone who's lying to you is generally regarded as being misled, lied to, or a lack of confirming the conclusions. Still totally different. I can hardly believe that the war models suggested by the AI actually led to a psychotic break with reality.@DejahEntendu @Rataunderground pretty sure it was more of a religious psychosis than an AI psychosis #JustSaying
-
@kcarruthers
I was just coming to say that I knew it wasn't your language. I'm sorry for yelling at you. Clearly I find the language problematic and I hate that people just accept it.@DejahEntendu IKR that is why it was in "quotes"
-
The trouble with a decapitation strike is if your enemy has broken itself down into multiple cells and has commands to 'keep following most recent commands until a next leadership issues new orders'.
@gnoll110 it is almost like they expected the Israelis to do that
-
@kcarruthers Just astounding. We really are in the Idiocracy timeline, aren't we?
Hesgeth didn't just leave all the war planning to a Palantir AI system based on Claude.
He insisted that all the guardrails were "woke" and must be switched off.
And he listened to Claude's analysis over the advice of experienced intelligence officers, who warned that removing the dictator was unlikely to bring down the regime, or that Iran could restrict oil supply.
From the article:
"Hegseth made his philosophy explicit at SpaceX in mid-January: “We will judge AI models on this standard alone: factually accurate, mission-relevant, without ideological constraints that limit lawful military applications. Department of War AI will not be woke.”
"The phrase “ideological constraints” performed significant rhetorical work. Safety filters designed to prevent AI hallucinations, sycophantic validation of flawed premises, and overconfident probability estimates were recast as political obstacles — “woke” limitations imposed by Silicon Valley liberals on America’s warfighters. The distinction between a guardrail that prevents an AI from endorsing a genocide and a guardrail that prevents an AI from inflating a strike’s predicted success rate was erased in a single speech.
...
"AI planning models assumed decapitating the senior leadership would trigger cascading institutional failure — IRGC fracture, Assembly of Experts deadlock, pro-reform street protests. Instead, the regime’s mosaic defence architecture — designed to survive exactly this scenario — activated as intended. By March 9, Mojtaba Khamenei had been installed as the new supreme leader, having reportedly survived the airstrike that killed his father by minutes.
"The National Intelligence Council had assessed beforehand that “even a large-scale assault would probably not collapse Iran’s clerical-military order.” This assessment had little impact. The polished, high-confidence outputs of AI simulation models proved more compelling to decision-makers than the hedged, probabilistic language of traditional intelligence analysis. As Foreign Policy noted, the succession “signals regime exhaustion” but not regime death — the IRGC backed continuity precisely because it was preferable to the chaos American planners expected to exploit."
https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlhf/@aj @kcarruthers Mind boggling.
-
@kcarruthers Just astounding. We really are in the Idiocracy timeline, aren't we?
Hesgeth didn't just leave all the war planning to a Palantir AI system based on Claude.
He insisted that all the guardrails were "woke" and must be switched off.
And he listened to Claude's analysis over the advice of experienced intelligence officers, who warned that removing the dictator was unlikely to bring down the regime, or that Iran could restrict oil supply.
From the article:
"Hegseth made his philosophy explicit at SpaceX in mid-January: “We will judge AI models on this standard alone: factually accurate, mission-relevant, without ideological constraints that limit lawful military applications. Department of War AI will not be woke.”
"The phrase “ideological constraints” performed significant rhetorical work. Safety filters designed to prevent AI hallucinations, sycophantic validation of flawed premises, and overconfident probability estimates were recast as political obstacles — “woke” limitations imposed by Silicon Valley liberals on America’s warfighters. The distinction between a guardrail that prevents an AI from endorsing a genocide and a guardrail that prevents an AI from inflating a strike’s predicted success rate was erased in a single speech.
...
"AI planning models assumed decapitating the senior leadership would trigger cascading institutional failure — IRGC fracture, Assembly of Experts deadlock, pro-reform street protests. Instead, the regime’s mosaic defence architecture — designed to survive exactly this scenario — activated as intended. By March 9, Mojtaba Khamenei had been installed as the new supreme leader, having reportedly survived the airstrike that killed his father by minutes.
"The National Intelligence Council had assessed beforehand that “even a large-scale assault would probably not collapse Iran’s clerical-military order.” This assessment had little impact. The polished, high-confidence outputs of AI simulation models proved more compelling to decision-makers than the hedged, probabilistic language of traditional intelligence analysis. As Foreign Policy noted, the succession “signals regime exhaustion” but not regime death — the IRGC backed continuity precisely because it was preferable to the chaos American planners expected to exploit."
https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlhf/@aj @kcarruthers
Following in the foosteps of McNamara.
Next up: Conscription with lowered fitness standards. -
@kcarruthers Just astounding. We really are in the Idiocracy timeline, aren't we?
Hesgeth didn't just leave all the war planning to a Palantir AI system based on Claude.
He insisted that all the guardrails were "woke" and must be switched off.
And he listened to Claude's analysis over the advice of experienced intelligence officers, who warned that removing the dictator was unlikely to bring down the regime, or that Iran could restrict oil supply.
From the article:
"Hegseth made his philosophy explicit at SpaceX in mid-January: “We will judge AI models on this standard alone: factually accurate, mission-relevant, without ideological constraints that limit lawful military applications. Department of War AI will not be woke.”
"The phrase “ideological constraints” performed significant rhetorical work. Safety filters designed to prevent AI hallucinations, sycophantic validation of flawed premises, and overconfident probability estimates were recast as political obstacles — “woke” limitations imposed by Silicon Valley liberals on America’s warfighters. The distinction between a guardrail that prevents an AI from endorsing a genocide and a guardrail that prevents an AI from inflating a strike’s predicted success rate was erased in a single speech.
...
"AI planning models assumed decapitating the senior leadership would trigger cascading institutional failure — IRGC fracture, Assembly of Experts deadlock, pro-reform street protests. Instead, the regime’s mosaic defence architecture — designed to survive exactly this scenario — activated as intended. By March 9, Mojtaba Khamenei had been installed as the new supreme leader, having reportedly survived the airstrike that killed his father by minutes.
"The National Intelligence Council had assessed beforehand that “even a large-scale assault would probably not collapse Iran’s clerical-military order.” This assessment had little impact. The polished, high-confidence outputs of AI simulation models proved more compelling to decision-makers than the hedged, probabilistic language of traditional intelligence analysis. As Foreign Policy noted, the succession “signals regime exhaustion” but not regime death — the IRGC backed continuity precisely because it was preferable to the chaos American planners expected to exploit."
https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlhf/@aj @kcarruthers yeah but look, in fairness to the poor hard done by codger, he rooly is learning & growing... this time he did not invite any journos into his secret chats...
-
Lolwut: “Was the Iran War Caused by AI Psychosis?
AI simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.”
Was the Iran War Caused by AI Psychosis? | House of Saud
AI sycophancy, RLHF bias, and Ender's Foundry simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.
House of Saud (houseofsaud.com)
@kcarruthers Not exactly a trustworthy source, but still: LOL
-
@C0rn3lius definitely.
-
Lolwut: “Was the Iran War Caused by AI Psychosis?
AI simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.”
Was the Iran War Caused by AI Psychosis? | House of Saud
AI sycophancy, RLHF bias, and Ender's Foundry simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.
House of Saud (houseofsaud.com)
-
@HydrePrever yup - but who can even trust the alleged ai checkers too
️ -
Lolwut: “Was the Iran War Caused by AI Psychosis?
AI simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.”
Was the Iran War Caused by AI Psychosis? | House of Saud
AI sycophancy, RLHF bias, and Ender's Foundry simulations shaped Operation Epic Fury. 7 planning assumptions failed in 23 days as the Iran war defied every AI prediction.
House of Saud (houseofsaud.com)
@kcarruthers #AI helped to derail the #FelonsCrusade in #Iran , starting with bombing the girls school in #Minab. https://www.independent.co.uk/news/world/americas/us-politics/iran-school-attack-ai-investigation-b2937456.html
But it was *started* by two felons who want to stay out of jail. https://newrepublic.com/post/207228/donald-trump-benjamin-netanyahu-iran
-
@kcarruthers Just astounding. We really are in the Idiocracy timeline, aren't we?
Hesgeth didn't just leave all the war planning to a Palantir AI system based on Claude.
He insisted that all the guardrails were "woke" and must be switched off.
And he listened to Claude's analysis over the advice of experienced intelligence officers, who warned that removing the dictator was unlikely to bring down the regime, or that Iran could restrict oil supply.
From the article:
"Hegseth made his philosophy explicit at SpaceX in mid-January: “We will judge AI models on this standard alone: factually accurate, mission-relevant, without ideological constraints that limit lawful military applications. Department of War AI will not be woke.”
"The phrase “ideological constraints” performed significant rhetorical work. Safety filters designed to prevent AI hallucinations, sycophantic validation of flawed premises, and overconfident probability estimates were recast as political obstacles — “woke” limitations imposed by Silicon Valley liberals on America’s warfighters. The distinction between a guardrail that prevents an AI from endorsing a genocide and a guardrail that prevents an AI from inflating a strike’s predicted success rate was erased in a single speech.
...
"AI planning models assumed decapitating the senior leadership would trigger cascading institutional failure — IRGC fracture, Assembly of Experts deadlock, pro-reform street protests. Instead, the regime’s mosaic defence architecture — designed to survive exactly this scenario — activated as intended. By March 9, Mojtaba Khamenei had been installed as the new supreme leader, having reportedly survived the airstrike that killed his father by minutes.
"The National Intelligence Council had assessed beforehand that “even a large-scale assault would probably not collapse Iran’s clerical-military order.” This assessment had little impact. The polished, high-confidence outputs of AI simulation models proved more compelling to decision-makers than the hedged, probabilistic language of traditional intelligence analysis. As Foreign Policy noted, the succession “signals regime exhaustion” but not regime death — the IRGC backed continuity precisely because it was preferable to the chaos American planners expected to exploit."
https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlhf/@aj @kcarruthers
Appears that Ai has won the first battle against humanity. -
R relay@relay.mycrowd.ca shared this topic
