Import AI 394: Global MMLU; AI safety needs AI liability; Canada backs Cohere

Import AI 393: 10B distributed training run; China VS the chip embargo; and moral hazards of AI development

Import AI 392: China releases another excellent coding model; generative models and robots; scaling laws for agents

Import AI 391: China's amazing open weight LLM; Fields Medalists VS AI Progress; wisdom and intelligence

Import AI 390: LLMs think like people; neural Minecraft; Google's cyberdefense AI

Import AI 389: Minecraft vibe checks; Cohere's multilingual models; and Huawei's computer-using agents

Import AI 388: Simulating AI policy; omni math; consciousness levels

Import AI 387: Overfitting vs reasoning; distributed training runs; and Facebook's new video models

Import AI 386: Google's chip-designing AI keeps getting better; China does the simplest thing with Emu3; Huawei's 8-bit data format

Import AI 385: False memories via AI; collaborating with machines; video game permutations

Import AI 384: Accelerationism; human bit-rate processing; and Google stuffs DOOM inside a neural network

Import AI 383: Automated AI scientists; cyborg jellyfish; what it takes to run a cluster