Groundbreaking research reveals that AI language models wastefully activate their entire networks for social reasoning tasks while actually using only a tiny specialized subset of parameters, creating massive inefficiency compared to human brains and pointing toward revolutionary energy-saving AI designs.
Nvidia CEO Jensen Huang warns China is winning the AI race despite U.S. chip bans, citing that 50% of global AI researchers and most leading open source models come from China while Beijing subsidizes energy costs for local alternatives.
Microsoft launches new MAI Superintelligence Team led by AI chief Mustafa Suleyman to develop advanced artificial intelligence applications for healthcare diagnostics, education, and renewable energy, marking the tech giant's push to reduce OpenAI dependence and compete with Meta's similar research investments.
Global data center electricity consumption races toward nearly doubling by 2030 as AI-driven servers surge with 30% annual growth, with the US and China commanding 80% of this massive energy expansion that will consume 3% of world electricity.
Chinese researchers at Peking University develop groundbreaking analog chip using resistive memory that delivers 1,000x faster performance than top Nvidia and AMD GPUs while consuming 100x less energy, solving precision issues that have plagued analog computing for decades.
University of Surrey scientists develop revolutionary brain-mimicking AI called Topographical Sparse Mapping that connects neurons only to nearby related ones, dramatically slashing energy consumption by over a million kilowatt-hours while boosting performance through biological pruning processes that refine connections during learning.
Scientists at University of Texas at Dallas unveil a revolutionary brain-inspired computer that learns like humans while consuming 90% less energy than current AI systems, potentially bringing powerful machine learning capabilities to mobile devices and dramatically reducing the environmental impact of artificial intelligence.
Amazon launches the world's largest AI data center in Indiana as part of its $11 billion Project Rainier, spanning 1,200 acres with 500,000 custom Trainium chips to train Anthropic's Claude models, while local residents worry about farmland loss and soaring utility costs.
Major technology companies are using non-disclosure agreements with local governments across 14 states to secretly develop AI data centers, preventing residents from knowing who is building in their communities and sparking revolts in places like Missouri and Arizona where transparency-starved citizens are enacting construction bans over concerns about utility costs …
OpenAI warns the White House that China's massive lead in power generation—adding 429 gigawatts versus America's 51 gigawatts last year—creates an 'electron gap' that threatens U.S. AI dominance, urging 100 gigawatts of annual capacity increases to compete.