AI Index Warning Statistics
Key point
AI performance has surged, but power, water, employment, and transparency burdens have grown right along with it.
Details
AI capability is rising fast, but environmental costs and opacity are growing just as fast.
- Grok 4's estimated training emissions were 72,816 tons CO2e, and AI data center power capacity grew to 29.6GW.
- Water usage for GPT-4o inference could exceed the annual drinking water demand of 12 million people.
- Frontier models have begun to surpass human-level performance on PhD-level science problems, multimodal reasoning, and competition math, but they remain weak at reading time, long-term planning, and sophisticated video generation.
The gap between the US and China has nearly disappeared. As of March 2026, Anthropic's top model held only a 2.7% edge, and the number of AI scholars moving to the US has fallen 89% compared to 2017. Global corporate AI investment surged to $581.7 billion in 2025, with private investment at $344.7 billion, and US AI investment reached $285.9 billion.
Early shocks are already visible in the labor market. Employment of software developers aged 22-25 has dropped nearly 20% since 2024, and the AI transparency index fell from 58 points to 40 points.
Generative AI reached 53% of the world's population within three years, and US consumer benefits are estimated at $172 billion annually. Globally, 59% of respondents are optimistic about AI's benefits, but the US had the lowest trust in regulation at 31%. While 80% of US high school and college students use AI for schoolwork, only half of schools have policies in place, and only 6% of teachers said their school's policy is clear.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.