Continuous Learning Delayed? AGI Breakthroughs Predicted in 12-24 Months
toptickcrypto · x · 2026-08-03
The author questions the progress of 'continuous learning' previously hinted at by AI labs, suggesting it might be delayed by 1-2 years similar to agent technology.
They predict that if either continuous learning or Recursive Self-Improvement (RSI) occurs within the next 12-24 months, it will trigger a massive industry bubble and blow-off top.
More from AGI Musings
- Steve Yegge Recommends Classic Sci-Fi: 'They're Made out of Meat' — Steve_Yegge · 2026-08-03
- Grok on Open Weights: Closed Labs Block, Openness Misbranded as Chinese Strategy — JosephJacks_ · 2026-08-03
- After Coldcard Hack, Capital Accumulation Seems Weak Defense Against ASI Disempowerment — MatthewMcAteer0 · 2026-08-03
- OpenAI Analyzes 800k Work Chats: AI Dissolves Traditional Job Boundaries — zainhas · 2026-08-03
- AI to Disproportionately Hit Junior Roles: Offline Skills Become the Moat — cto_junior · 2026-08-03
- AI Math Advances Shrouded in Secrecy, Scientists Urge Open Methods — rbhar90 · 2026-08-03