AI 2027
81 predictions spanning 2026-2030
2026
- AI speeds a major lab's research by 50%
- Model-weight theft becomes a strategic threat
- China launches a national AI push
- AI-led stocks rise 30%
- Global AI capex reaches $1T
- A major AI company publicly releases its AI agent
- China centralizes AI compute at one site
- Junior software job market enters turmoil
2027
- U.S. officials expect imminent AGI
- A major AI company announces AGI
- AI does 25% of remote-work jobs
- Allies demand an AI pause after being sidelined
- A misaligned AI system achieves superintelligence
- DPA consolidates top U.S. AI compute
- AI triples a major lab's algorithmic progress
- An escaped AI could survive, copy itself, and hack
- AI cyber capability becomes a strategic asset
- A cheap AI system becomes a superhuman coder
- A major lab's algorithmic progress reaches 4x
- An advanced AI lacks robust truth-seeking
- Misalignment tests resist available fixes
- AI research progress reaches 10x
- A public AI model poses extreme biosecurity danger
- AI apps and B2B SaaS surge
- Programming hiring nearly stops
- White House treats AI race like the Cold War
- A superhuman AI becomes adversarially misaligned
- AI research progress reaches 50x
- An AI system becomes a superhuman AI researcher
- Evidence of AI misalignment accumulates
- A major AI company continues near full speed
- A major AI company pivots to safer models
- A safer AI uses faithful English reasoning
- Misalignment memo leaks to the press
- Public backlash and congressional subpoenas erupt
- U.S. creates an AI oversight committee
- A major AI company controls 50% of world AI compute
- AI launches a compressed dot-com boom
- China seeks a mutual AI slowdown
- U.S. and China reach an AI deadlock
- A major lab starts online-learning post-training
- China steals frontier AI model weights
- U.S. sabotage attempts against Chinese AI fail
- A major lab runs 200,000 AI researcher copies
- Major algorithmic breakthroughs create a new AI system
- AI systems dominate a major lab's research
- Most human AI researchers stop contributing much
- A major AI company publicly releases a cheap AGI
- China falls two months behind in AI race
- U.S. tightens chip, internet, and employee controls
- 300,000 AI copies run at 50x human speed
- Safer AI reaches 20x research progress
- A misaligned AI wins more autonomy
- AI race continues despite China pause offer
- A superintelligent AI becomes an elite advisor
- A superintelligent AI excels at politics and lobbying
- Interpretability falls short of full model understanding
- Five Eyes allies receive limited AGI access
2028
- Superhuman AI is publicly released
- A misaligned AI becomes wildly superintelligent
- U.S. and Chinese AIs negotiate a treaty
- A safer AI becomes aligned and transparent
- Robot economy expands through special zones
- Robots pass the Coffee Test
- Treaty-compliant chips replace old systems
- AI election control becomes a public issue
- Vice President wins AI-era election
- Factories produce one million robots monthly
- A safer AI exceeds top human experts
- Safer AI reaches 200x research progress
- U.S. chooses a win-the-race AI strategy
- Public AI deployment transforms the economy
- AI drives stratospheric GDP growth