As AI chatbots become an integral part of digital life, their environmental impact is coming under scrutiny. A recent analysis by TRG Datacenters reveals massive differences in carbon emissions among major AI models, with Grok AI emerging as the most eco-friendly option.
Grok AI Tops the Sustainability Charts
According to the study, Grok AI registers a minuscule carbon footprint of just 0.17 grams of CO₂ per query, making it the most environmentally responsible chatbot in the industry. This ultra-low energy consumption is attributed to its optimised architecture, which minimises computational demands while maintaining functionality. In terms of impact, a single Grok AI query generates about the same emissions as a Google search — an impressive feat for AI efficiency.
In contrast, Google Gemini, which ranks second, emits 1.6g CO₂ per query — nearly ten times higher than Grok. Despite this, Gemini benefits from Google’s renewable energy investments, making it significantly less polluting than many other competitors.
How Other AI Models Compare
Meta’s LLaMA AI comes in third, producing 3.2g CO₂ per query. While Meta has made strides in sustainability, its expanding AI operations could push emissions even higher in the future. The emissions from using LLaMA are roughly equivalent to sending 10 emails.
| AI Model | Estimated Inference CO₂ per Query |
| Grok AI | 0.17 |
| Google Gemini | 1.6 |
| LLaMA (Meta AI) | 3.2 |
| Claude AI | 3.5 |
| Perplexity AI | 4 |
| ChatGPT (GPT-4) | 4.32 |
Claude AI, ranking fourth, records 3.5g CO₂ per query. While it prioritises safety and reliability, these features seem to increase computational load, leading to a moderate environmental footprint.
Perplexity AI takes the fifth spot with 4g CO₂ per query, largely due to its integrated search functionalities that require additional processing power. The model’s carbon impact is comparable to charging a smartphone one to 1.5 times.
GPT-4: The Biggest Polluter in the AI Race
At the bottom of the sustainability rankings sits GPT-4, which emits a staggering 4.32g CO₂ per query—25 times more than Grok AI. As one of the most advanced AI models, GPT-4 requires vast computational resources, translating to a significant environmental cost. Each interaction with GPT-4 is comparable to sending 21 emails or nearly a full phone charge, raising concerns about the long-term sustainability of such high-power AI systems.
The Path to Sustainable AI
A spokesperson from TRG Datacenters commented on the findings, stating, “As AI adoption continues to rise, finding ways to reduce its energy consumption will be key. Some models are already designed to be more efficient, but there is still room for improvement. Advances in hardware, more optimised AI models, and increased use of renewable energy in data centers could help lower emissions over time. AI is here to stay, but balancing innovation with sustainability will be essential in minimising its environmental impact.”
The study underscores the urgent need for AI companies to prioritise sustainability, as computational demands continue to rise. With models like Grok AI proving that efficiency is possible, the pressure is now on other tech giants to follow suit.
