Can AI self-improvement overcome diminishing returns?

kept by eddie sep 28

Despite rapid AI progress, the self‑improvement feedback loop is too weak to overcome diminishing returns, making a rapid intelligence explosion unlikely without a major breakthrough.

Before he wrote AI 2027, he predicted the world in 2026. How did he do?
Daniel Kokotajlo's 2021 narrative predictions for 2026 were largely accurate on AI revenue, agent scaffolds, and assistant adoption, but overestimated chip fab speed and AI-driven political polarization.
also on AI 2027, #forecasting, #ai-research, OpenAI, Anthropic, #ai
Scenarios for our Economic Future \ Anthropic
Anthropic's economic model of 2030 shows AI boosts GDP across all scenarios, but in extreme cases, knowledge-worker wages fall over 10% and unemployment spikes, with capital capturing most gains.
also on #forecasting, Anthropic, #automation
Are Prediction Markets Good for Anything?—Asterisk
Despite billions in volume, Polymarket and Kalshi prediction markets produce useful forecasts mainly on high-profile risks, while AI chatbots may soon replace them as the primary source of probabilistic information.
also on #forecasting, #ai
400 Bad Request
AGI capable of most cognitive work could arrive by 2028–2034, but deployment will lag capability due to verification bottlenecks, uneven automation, and institutional friction.
also on METR, #ai-research, #automation, #ai
The Shape of the Thing - by Ethan Mollick
AI has entered a new agentic era where models can autonomously complete hours of human work, and exponential capability improvements are now driving radical organizational experiments like StrongDM's "Software Factory" that ships code without human touching it.
also on METR, #ai-research, Anthropic