Tag

Articles tagged: research

2 articles

News July 5, 2026

OpenAI's Original Scaling Law Had a Bug That Led the Industry to Build Oversized, Undertrained Models for Two Years

Former OpenAI researcher Diogo Almeida published a detailed breakdown of the bug in OpenAI's original 2020 scaling laws, confirming that a combination of fixed training data, an aggressive learning rate decay, and a misleading independence claim caused the entire AI industry to build oversized, undertrained models until DeepMind's Chinchilla correction in 2022.

News April 28, 2026

3 min read

DeepMind's David Silver Raises $1.1 Billion Seed Round at $5.1 Billion Valuation for Reinforcement Learning Lab Ineffable Intelligence

David Silver, the reinforcement learning researcher behind AlphaGo, has raised $1.1 billion in seed funding for Ineffable Intelligence at a $5.1 billion valuation. The London-based lab aims to build AI 'superlearners' that discover knowledge through trial and error rather than training on human data, backed by Sequoia, Lightspeed, Nvidia, Google, and the UK Sovereign AI Fund.

← Back to all stories