OpenAI’s GPT-5 draws fire for massive energy use
OpenAI rolled out GPT-5 on Thursday, prompting experts to flag its huge power consumption. The new AI reportedly uses 5 to 20 times more electricity to generate responses than its predecessors.
Researchers at the University of Rhode Island found GPT-5 can guzzle up to 40 watt-hours of power for a medium-length reply—roughly the energy an incandescent bulb burns in 40 minutes. Their dashboard shows an average of over 18 watt-hours per response, exceeding all other benchmarked models except a couple of recent heavyweights. 
This is a sharp jump from GPT-4o’s consumption. Given ChatGPT cranks through 2.5 billion daily prompts, GPT-5’s total energy use could rival the daily electricity consumption of 1.5 million US homes.
OpenAI CEO Sam Altman gave vague resource figures earlier this year but hasn’t published official, detailed power metrics since GPT-3 in 2020. Experts say GPT-5’s advanced reasoning, image and video capabilities, plus its vast model size (believed to be multiple times larger than GPT-4), inflate its environmental footprint.
University of Illinois professor Rakesh Kumar explained:
“A more complex model like GPT-5 consumes more power both during training and during inference. It’s also targeted at long thinking … I can safely say that it’s going to consume a lot more power than GPT-4.”
The University of California’s Shaolei Ren also warned reasoning mode increases compute time by 5 to 10 times.
“If you use the reasoning mode, the amount of resources you spend for getting the same answer will likely be several times higher, five to 10,” said Ren.
The University of Rhode Island team saw major hurdles estimating actual power draw due to lack of data from AI companies. They stress the need for transparency:
“It’s more critical than ever to address AI’s true environmental cost,” said URI’s Marwan Abdelatti.
“We call on OpenAI and other developers to use this moment to commit to full transparency by publicly disclosing GPT-5’s environmental impact.”
The issue started with GPT-5’s launch showcasing breakthrough capabilities like website creation and PhD-level reasoning, but now energy costs are pulling attention back.
OpenAI has yet to publicly reveal GPT-5’s parameter count or detailed environmental stats. For now, the world knows GPT-5’s power appetite has just hit new highs—and questions over AI’s carbon footprint keep coming.
