Should we feel guilty about the climate impact of AI?

My own conclusion after doing these calculations is that it’s not small… but it’s also not big enough to make 𝘵𝘩𝘦 𝘢𝘷𝘦𝘳𝘢𝘨𝘦 𝘤𝘪𝘵𝘪𝘻𝘦𝘯 feel guilty about it. I think it’s important to have a public discussion on AI (the environmental impact might not be the biggest issue). But actually, this video aims to answer a different question: which kind of 𝗯𝗲𝗵𝗮𝘃𝗶𝗼𝘂𝗿𝗮𝗹 𝗰𝗵𝗮𝗻𝗴𝗲𝘀 should we campaign for? Individual behaviour change is not the solution to all climate problems, but it’s definitely a significant part of the solution. Take the Jump puts the number at 25 percent: a quarter of all necessary emissions savings in rich countries needs to come from individual behaviour change. And almost half of that is related to food. People have limited attention and willpower. Confronting people with the carbon footprint of everything is confusing and paralyzing. So with Fork Ranger, our goal is to show where individual behaviour change really matters. And food is a huge part of it. —-— SOURCES 💻 For the AI footprint I used the upper estimate of this study: de Vries-Gao, A. (2026). The carbon and water footprints of data centers and what this could mean for artificial intelligence. *Patterns*, *7*(1). ⚽️ World cup: South Pole (South Pole Carbon Asset Management Ltd.), SETTERWALL S., ASTAD (2021/06). Greenhouse gas accounting report. FIFA World Cup 2022 🥩 Beef footprint: Xu, X., Sharma, P., Shu, S., Lin, T. S., Ciais, P., Tubiello, F. N., ... & Jain, A. K. (2021). Global greenhouse gas emissions from animal-based foods are twice those of plant-based foods. Nature food, 2(9), 724-732. They calculated a total footprint of beef of around 4300 million tons CO2e. So Ai is 1.86% of beef’s total footprint. I then used beef production numbers from FAO for 2024: 77 million tons per year. 1.86% of that, divided by 8,15 billion people is ~175 grams or roughly one steak per person per year.