The Alignment Research Center - Millicosm

kept by novawanderer sep 26

ARC proposes that any surprising mathematical fact has a heuristic explanation, aiming to use this to align future AI systems.

Shtetl-Optimized » Blog Archive » The Age of Wonders and Terrors
The AI singularity has begun, as AI now solves major open math problems like Navier‑Stokes, proving it is the primary theorem‑prover and demanding acceptance of its reality.
also on Alignment Research Center, GPT-6, #machine-learning, Anthropic, #ai
What is neuralese and why is it bad? — LessWrong
Neuralese refers to models' internal non‑linguistic thought vectors, which threaten interpretability by hiding intent from chain‑of‑thought monitoring and complicating safety oversight.
also on Alignment Research Center, #machine-learning, Anthropic, #ai
Why I don't think AI is a bubble
The combination of large language models with reinforcement learning creates a path for continued improvement, making the argument that AI progress will plateau unlikely.
also on GPT-3, #machine-learning, #ai
The plunging price of thought
Epoch AI reports that the cost to achieve a given level of AI performance has fallen roughly 47% per quarter since 2023, outpacing historic tech price drops.
also on GPT-3, #ai
OpenAI did it again. But what did they do? - Ben Hylak
OpenAI’s newly revealed Computer Use feature lets AI perform general computer tasks, marking the sixth step toward AGI and threatening to eliminate human gap‑filler roles.
also on GPT-6