1980s AI photo trend: What is the environmental cost of your latest AI avatar?
The viral 80s AI trend is a social media craze in which users are using ChatGPT and Gemini to reimagine their modern selfies as retro portraits from the 1980s.
In the 1980s, a new trend emerged: AI-generated images. Yet, the environmental impact of these digital creations may come as a surprise. Every AI-generated image requires computational power, which consumes electricity. One AI image may seem innocuous, but as millions of people join the trend and create multiple versions, the collective impact grows.
Research from Carnegie Mellon University and Hugging Face suggests that generating an AI image uses between 0.01 and 0.29 kilowatt-hours of electricity, depending on the system and process. This amount is enough to power a 10-watt LED bulb for about 17 minutes. For a complex AI video, the energy required could power the same bulb for 42 hours.
The majority of AI energy usage happens during inference, or the day-to-day use of AI models. A popular AI platform like ChatGPT processes around 2.5 billion prompts daily, consuming hundreds of gigawatt-hours of electricity annually. Model choice, prompt length, output format, and resolution all influence the energy footprint, but users often have no visibility into these decisions.
While efficiency improvements may help, the rebound effect could increase total resource consumption. A study predicts that AI infrastructure's global water footprint could reach 4.2–6.6 billion cubic meters annually by 2027, with two-thirds of data centers located in water-stressed regions. Additionally, AI could generate up to 2.5 million tonnes of electronic waste and consume 9.3 trillion liters of water by 2030.
Environmentalists have raised concerns about Google's planned $15 billion data-center hub in Andhra Pradesh, citing worries about water supplies and wildlife. The project faces opposition due to its location near a wildlife sanctuary and potential impacts on Visakhapatnam's already strained water resources.
Written by urgent.news from Hindustan Times's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
