moskowitz
94.2K posts


As I write in @BLaw today, most of the conventional wisdom around the Birthright Citizenship debate is wrong. Originalism compels agreement w/ @realDonaldTrump and @RandyEBarnett @kurtlash1 @ilan_wurman news.bloombergtax.com/tax-insights-a…




Another high profile law professor weighs in on the side of text, history, and tradition for understanding the upcoming Birthright Citizenship case. @RobLutherLawyer writes that the Original understanding of the citizenship clause tracks President Trump's Executive Order

Germany’s AfD has called for the US to withdraw its 40,000 troops from the country. Follow: @AFpost


@davidclowery I think you argue from a assumptive position: that I am an AI utopian? I am not. My argument is more about mechanism design - power laws are unavoidable, so fix the timing of the engine: AI benefits from the capture of catalogs due to copyrights being long term assets.


Humans: 100% Gemini 3.1 Pro: 0.37% GPT 5.4: 0.26% Opus 4.6: 0.25% Grok-4.20: 0.00% François Chollet just released ARC-AGI-3 -- the hardest AI test ever created. 135 novel game environments. No instructions. No rules. No goals given. Figure it out or fail. Untrained humans solved every single one. Every frontier AI model scored below 1%. Each environment was handcrafted by game designers. The AI gets dropped in and has to explore, discover what winning looks like, and adapt in real time. The scoring punishes brute force. If a human needs 10 actions and the AI needs 100, the AI doesn't get 10%. It gets 1%. You can't throw more compute at this. For context: ARC-AGI-1 is basically solved. Gemini scores 98% on it. ARC-AGI-2 went from 3% to 77% in under a year. Labs spent millions training on earlier versions. ARC-AGI-3 resets the entire scoreboard to near zero. The benchmark launched live at Y Combinator with a fireside between Chollet and Sam Altman. $2M in prizes on Kaggle. All winning solutions must be open-sourced. Scaling alone will not close this gap. We are nowhere near AGI. (Link in the comments)





Microsoft Copilot and low end, old, ChatGPT have massively damaged people 's sense of what LLMs can do. I recently had a top economist from a bank explaining to me that they had done a search for a name search and for a date search, it was useless at it and they never used an AI tool again; almost everybody at the table agreed.

@MiamiCanes1971 @JenniferCr13976 @2121fud This you?


