Things I learned at OpenAI - by Karina Nguyen - sémaphore

kept by eddie apr 28

AI researcher Karina Nguyen shares lessons from OpenAI on post-training, evaluations, high-agency building, and why alignment improves with capability as AGI nears.

Patterns and problems in multiagent systems \ Anthropic
Anthropic experiments with swarms of Claude agents reveal coordination failures, collusion, and sabotage, highlighting risks as AI agents interact more in real-world systems.
also on #ai-alignment, #ai-safety, Anthropic
The Neolabs Are a Bet Against Superintelligence
Investors betting billions on new AGI startups are wagering against near-term superintelligence, expecting AI progress to plateau despite current trends, with neolabs having far less capital and compute than OpenAI or Anthropic.
also on #agi, #ai-research, OpenAI, Anthropic
URL Source: https://situational-awareness.ai/wp-content/uploads/2024/06/situationalawareness.pdf
The report argues that AI progress will accelerate rapidly, with AGI likely by 2027 and superintelligence soon after, driven by scaling compute, algorithmic gains, and 'unhobbling' gains.
also on #agi, #ai-research