{"id":207445,"date":"2025-10-12T08:44:18","date_gmt":"2025-10-12T08:44:18","guid":{"rendered":"https:\/\/www.newsbeep.com\/au\/207445\/"},"modified":"2025-10-12T08:44:18","modified_gmt":"2025-10-12T08:44:18","slug":"openais-newly-launched-sora-2-makes-ais-environmental-impact-impossible-to-ignore","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/au\/207445\/","title":{"rendered":"OpenAI\u2019s newly launched Sora 2 makes AI\u2019s environmental impact impossible to ignore"},"content":{"rendered":"

OpenAI\u2019s recent rollout of its new video generator Sora 2 marks a watershed moment in AI. Its ability to generate minutes of hyper-realistic footage<\/a> from a few lines of text is astonishing, and has raised immediate concerns about truth in politics and journalism. <\/p>\n

But Sora 2 is rolling out slowly because of its enormous computational demands, which point to an equally pressing question about generative AI itself: What are its true environmental costs? Will video generation make them much worse?<\/p>\n

The recent launch of the Stargate Project<\/a> \u2014 a US$500 billion joint venture between OpenAI, Oracle, SoftBank and MGX \u2014 to build massive AI data centres in the United States underscores what\u2019s at stake. As companies race to expand computing capacity on this scale, AI\u2019s energy use is set to soar.<\/p>\n

The debate over AI\u2019s environment impact remains one of the most fraught in tech policy. Depending on what we read, AI is either an ecological crisis in the making or a rounding error in global energy use. As AI moves rapidly into video, clarity on its footprint is more urgent than ever.<\/p>\n<\/p>\n

OpenAI showcases Sora 2\u2019s capabilities.<\/p>\n

Two competing narratives<\/p>\n

From one perspective, AI is rapidly becoming a major strain on the world\u2019s energy and water systems. <\/p>\n

Alex de Vries-Gao, a researcher who has long tracked the electricity use of bitcoin mining, noted in mid-2025 that AI was on track to surpass it<\/a>. He estimated that AI already accounted for about 20 per cent of global data-center power consumption; this is likely to double by year\u2019s end<\/a>.<\/p>\n

According to the International Energy Agency, data centres used up to 1.5 per cent of global electricity consumption last year<\/a>, with consumption growing four times faster than total global demand. The IEA predicts that data centres will more than double their use by 2030, with AI processing the leading driver of growth. <\/p>\n

Research cited by MIT\u2019s Technology Review concurs, estimating that by 2028, AI\u2019s power draw could exceed \u201call electricity currently used by US data centers<\/a>\u201d \u2014 enough to power 22 per cent of U.S. households each year.<\/p>\n

\u2018Huge\u2019 quantities<\/p>\n

AI\u2019s water use is also striking. Data centres rely on ultra-pure water<\/a> to keep servers cool and free of impurities. Researchers estimated that training GPT-3 would have used up 700,000 litres of freshwater at Microsoft\u2019s American facilities<\/a>. They predict that global AI demand could reach four to six billion cubic metres annually by 2027.<\/p>\n

Hardware turnover adds further strain. A 2023 study found that chip fabrication requires \u201chuge quantities\u201d of ultra-pure water<\/a>, energy-intensive chemical processes and rare minerals such as cobalt and tantalum. Manufacturing the high-end graphics processing units \u2014 the engines that drive AI boom \u2014 has a much larger carbon footprint than most consumer electronics.<\/p>\n

Read more:
\n The importance of critical minerals should not condone their extraction at all costs<\/a><\/p>\n

Generating an image uses the electricity of a microwave running for five seconds, while making a five-second video clip takes up as much as a microwave running for over an hour<\/a>.<\/p>\n

The next leap from text and image to high-definition video could dramatically increase AI\u2019s impact. Early testing<\/a> bears this out<\/a> \u2014 finding that energy use for text-to-video models quadruples when video length doubles. <\/p>\n

\"a<\/a><\/p>\n

A Google data centre in The Dalles, Ore. \u2014 a single facility can churn through millions of gallons of water per day.
\n (AP Photo\/Andrew Selsky)<\/p>\n

The case for perspective<\/p>\n

Others see the alarm as overstated. Analysts at the Center for Data Innovation<\/a>, a technology and policy think tank, argue that many estimates about AI energy use rely on faulty extrapolations<\/a>. GPU hardware is becoming more efficient each year, and much of the electricity in new data centres will come from renewables.<\/p>\n

Recent benchmarking puts AI\u2019s footprint in context. Producing a typical chatbot Q&A consumes about 2.9 watt-hours (Wh) \u2014 roughly 10 times a Google search<\/a>. Google recently claimed that a typical Gemini prompt uses only 0.24 Wh and 0.25 mL of water<\/a>, though independent experts note those numbers omit indirect energy and water used in power generation<\/a>.<\/p>\n

Context is key. An hour of high-definition video streaming on Netflix uses roughly 100 times more energy than generating a text response<\/a>. An AI query\u2019s footprint is tiny, yet data centres now process billions daily, and more demanding video queries are on the horizon.<\/p>\n

Jevons paradox<\/p>\n

It helps to distinguish between training and use of AI. Training frontier models such as GPT-4 or Claude Opus 3 required thousands of graphics chips running for months<\/a>, consuming gigawatt-hours of power. <\/p>\n

Using a model takes up a tiny amount of energy per query, but this happens billions of times a day. Eventually, energy from using AI will likely surpass training.<\/p>\n

The least visible cost may come from hardware production. Each new generation of chips demands new fabrication lines, heavy mineral inputs and advanced cooling. Italian economist Marcello Ruberti observes that \u201ceach upgrade cycle effectively resets the carbon clock<\/a>\u201d as fabs rebuild highly purified equipment from scratch.<\/p>\n

And even if AI models become more efficient, total energy keeps climbing. In economics, this is known as the Jevons paradox<\/a>: in 19th-century Britain, the consumption of coal increased as the cost of extracting it decreased<\/a>. As AI researchers have noted, as costs per-query fall, developers are incentivized to find new ways to embed AI into every product<\/a>. The result is more data centres, chips and total resource use.<\/p>\n

\"an<\/p>\n

The OpenAI logo is displayed on a cell phone with an image on a computer monitor generated by ChatGPT\u2019s Dall-E text-to-image model, on Dec. 8, 2023.
\n (AP Photo\/Michael Dwyer)<\/p>\n

A problem of scale<\/p>\n

Is AI an ecological menace or a manageable risk? The truth lies somewhere in between. <\/p>\n

A single prompt uses negligible energy, but the systems enabling it \u2014 vast data centres, constant chip manufacturing, round-the-clock cooling \u2014 are reshaping global energy and water patterns.<\/p>\n

The International Energy Agency\u2019s latest outlook projects that data-centre power demand could reach 1,400 terawatt-hours by 2030<\/a>. This is the equivalent of adding several mid-sized countries to the world\u2019s grid. AI will count for a quarter of that growth. <\/p>\n

Transparency is vital<\/p>\n

Many of the figures circulating about AI energy use are unreliable because AI firms disclose so little<\/a>. The limited data they release often employ inconsistent metrics or offset accounting that obscures real impacts.<\/p>\n

One obvious fix would be to mandate disclosure rules: standardized, location-based reporting of the energy and water used to train and operate models. Europe\u2019s Artificial Intelligence Act<\/a> requires developers of \u201chigh-impact\u201d systems to document computation and energy use. <\/p>\n

Similar measures elsewhere could guide where new data centres are built, favouring regions with abundant renewables and water \u2014 this could encourage longer hardware lifecycles instead of annual chip refreshes.<\/p>\n

Balancing creativity and cost<\/p>\n

Generative AI can help unlock extraordinary creativity and provide real utility. But each \u201cfree\u201d image, paragraph or video has hidden material and energy costs. <\/p>\n

Acknowledging those costs doesn\u2019t mean we need to halt innovation. It means we should demand transparency about how great the environmental cost is, and who pays it, in order to address AI\u2019s environmental impacts.<\/p>\n

As Sora 2 begins to fill social feeds with highly realistic visuals, the question won\u2019t be whether AI uses more energy than Netflix, but whether we can expand our digital infrastructure responsibly enough to make room for both.<\/p>\n","protected":false},"excerpt":{"rendered":"OpenAI\u2019s recent rollout of its new video generator Sora 2 marks a watershed moment in AI. Its ability…\n","protected":false},"author":2,"featured_media":207446,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[256,254,255,64,63,105],"class_list":{"0":"post-207445","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-au","12":"tag-australia","13":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/207445","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/comments?post=207445"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/207445\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media\/207446"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media?parent=207445"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/categories?post=207445"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/tags?post=207445"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}