Another Monday Message Board. Post comments on any topic. Civil discussion and no coarse language please. Side discussions and idees fixes to the sandpits, please.
I’m now using Substack as a blogging platform, and for my monthly email newsletter. For the moment, I’ll post both at this blog and on Substack. You can also follow me on Mastodon here.
All this because Dems in the Midwest didn’t feel like voting for a woman. (Technically, I am guessing, about the motives. Lazy fckers.)
I don’t even know what to say.
I asked Google AI this rather clumsy question.
The answer was:
“Mathematical, logical, and strategic AI models—distinct from Large Language Models (LLMs)—that simulate the Iran conflict as of early 2026 generally predict a high-intensity, algorithmically-driven war characterized by rapid strikes and significant risks of escalation.
These systems, used for target identification and war-gaming, emphasize that the conflict’s outcome is heavily dependent on maintaining the, often fragile, “sensor-decision-shooter” chain, with a high probability of initial military success for a US-Israel coalition followed by long-term strategic uncertainty.
Key predictions and outcomes forecasted by strategic AI and game theory models include:
1. High Probability of Rapid Escalation (95% Scenario)
In simulations designed to mirror intense geopolitical crises, AI models have shown a tendency toward extreme outcomes, with a 95% likelihood of choosing rapid, high-intensity escalation over de-escalation.
2. Predicted Outcomes of Conflict
3. Key Factors Shaping the Math
Summary of Predicted Scenarios (as of March 2026)
Note: These predictions are based on operational data from late February to early March 2026, including the “Epic Fury” operations, which marked a transition to AI-mediated, high-speed warfare.”
That’s the end of the AI answer to an admittedly clumsy question. I won’t say much except that the answer appears to me to somewhat conflate or blur the issues of how pure AI would run the war as opposed to how humans using self-interest considerations, politics and AI would run the war. Or maybe I am jumping at shadows there.
Also, the answer’s links (not shown above) do seem to show a heavy reliance on journalistic reporting of issues and journalistic opinions (like Guardian articles and BBC reporting). It’s interesting that it did not use more academic or war college etc. sources in the public domain.
One could pursue this and refine questions in an attempt to clarify/redress this. I am not sure I want to go that far down the rabbit hole.
Footnote to above, my clumsy question to the AI was:
What do mathematical-logical-deductive-strategic AIs (not LLMs) that game out the Iran war predict for the result(s) of the war? Give probabilities for outcomes if necessary.