{"id":331741,"date":"2025-12-05T09:03:15","date_gmt":"2025-12-05T09:03:15","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/331741\/"},"modified":"2025-12-05T09:03:15","modified_gmt":"2025-12-05T09:03:15","slug":"the-rise-of-ai-reasoning-models-comes-with-a-big-energy-tradeoff","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/331741\/","title":{"rendered":"The Rise of AI Reasoning Models Comes With A Big Energy Tradeoff"},"content":{"rendered":"<p class=\"ArticleBodyText_articleBodyContent__17wqE typography_articleBody___5jDr\" data-component=\"paragraph\">Nearly all leading artificial intelligence developers are focused on building AI models that mimic the way humans reason, but new research shows these cutting-edge systems can be far more energy intensive, adding to concerns about AI\u2019s strain on power grids.<\/p>\n<p class=\"ArticleBodyText_articleBodyContent__17wqE typography_articleBody___5jDr\" data-component=\"paragraph\">AI reasoning models used 100 times more power on average to respond to 1,000 written prompts than alternatives without this reasoning capability or which had it disabled, according to a study released Thursday. The work was carried out by the AI Energy Score project, led by Hugging Face research scientist Sasha Luccioni and Salesforce Inc. head of AI sustainability Boris Gamazaychikov.<\/p>\n<p>The researchers evaluated 40 open, freely available AI models, including software from OpenAI, Alphabet Inc.\u2019s Google and Microsoft Corp. Some models were found to have a much wider disparity in energy consumption, including one from Chinese upstart DeepSeek. A slimmed-down version of DeepSeek\u2019s R1 model used just 50 watt hours to respond to the prompts when reasoning was turned off, or about as much power as is needed to run a 50 watt lightbulb for an hour. With the reasoning feature enabled, the same model required 308,186 watt hours to complete the tasks.<\/p>\n","protected":false},"excerpt":{"rendered":"Nearly all leading artificial intelligence developers are focused on building AI models that mimic the way humans reason,&hellip;\n","protected":false},"author":2,"featured_media":331742,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,13564,1540,181,507,2254,11637,2854,2075,1536,1538,79,7271,74],"class_list":["post-331741","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-alphabet-inc-cl-a","tag-amazon-com-inc","tag-artificial-intelligence","tag-artificialintelligence","tag-august","tag-chief-executive-officer","tag-green","tag-hardware","tag-microsoft-corp","tag-salesforce-inc","tag-science","tag-software","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/331741","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=331741"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/331741\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/331742"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=331741"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=331741"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=331741"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}