Meta's AI Ad Ambitions Fuel Climate Crisis with Fossil Gas

What Happened

Meta, the social media giant, is doubling down on artificial intelligence to supercharge its advertising business. This move comes as the company continues to rake in massive profits, with advertising accounting for a staggering 98% of its total revenue in the second quarter of 2025. In a recent earnings call, Meta executives proudly announced the growth of their "Advantage+" ad product and a 20% quarter-over-quarter increase in advertisers using their video generation features.

The vision is clear: these machine-learning tools will enable unprecedented levels of personalized marketing. By automatically churning out countless variations of images and videos, they plug directly into Meta's vast trove of user data. This means companies can fire marketing employees, replacing them with algorithms that brute-force ad creation, flooding users with tailored content designed to make them buy more stuff.

Why It Matters

Here's where the shiny AI facade begins to crack. Generating large amounts of digital content, especially video, is incredibly energy-intensive. A recent study revealed that video diffusion is roughly 30 times more costly than image generation, 2,000 times more than text generation, and a shocking 45,000 times more than text classification in terms of energy. Expanding video dimensions only increases this cost quadratically, leading to rapidly escalating hardware and environmental burdens.

Meta's own sustainability report shows a fivefold increase in electricity consumption since 2018. This directly corresponds to a rise in power grid emissions. While the company claims to offset this by purchasing renewable energy certificates, critics argue these certificates are often questionable, masking a significant environmental footprint. Worse, Meta has become an industry leader in building new fossil gas infrastructure, partnering with utilities like Entergy — which expects to miss its 2030 climate targets — to construct over two gigawatts of new fossil gas capacity.

The Environmental Toll

The timing of Meta's AI-driven expansion couldn't be worse for the planet. The United States is currently pouring money into resuscitating coal and accelerating gas production, even seeing an increase in coal generation for the first time in half a decade. Alarmingly, the U.S. has now surpassed China as the top global developer of new fossil gas power stations, with 37% of recent capacity linked to data centers and AI infrastructure. The Trump administration's "permitting reform" further enables tech companies to build their own fossil fuel plants, making Meta's actions part of a larger, concerning trend.

This isn't just about Meta's direct emissions. The mass production of sensory triggers designed to push consumption contributes to the accelerating climate crisis. Excessive online purchases and discards are already major environmental contributors. Now, Meta's push into the most energy-intensive forms of generative software, which users cannot easily boycott or turn off, promises to further boost the fossil fuel industry.

Ethical and Societal Concerns

Beyond the environmental impact, the quality and ethics of AI-generated content are deeply problematic. Automated systems often produce results that are nonsensical, racist, false, or even violent. Meta's own terms for its generative AI tools warn that outputs could be "inaccurate, incomplete, misleading, offensive and/or inappropriate." One company, for instance, saw Meta's software replace a top-performing ad with an "AI-generated photo of a cheerful yet unnatural granny sitting in an armchair."

This suggests the "secret sauce" isn't intelligent marketing, but rather a brute-force approach to mass-produce vast quantities of content, hoping something sticks. This raises serious questions about the moral necessity of such immense energy demands. Do we need billions of AI-generated ads more than we need precious electricity for hospitals, electrification, or climate disaster relief?

Our Take

Meta's strategy mirrors historical precedents where industries fabricated necessity to extend the life of fossil fuels – think of the plastics industry or car manufacturers bloating vehicle models. While Meta's advertising revenue remains eye-wateringly profitable, the path it's taking with generative AI comes with a significant societal and environmental price tag. The company's reliance on questionable renewable energy certificates and its direct investment in fossil gas infrastructure highlight a stark contradiction between its sustainability claims and its operational reality.

The carbon footprint per Meta user has already risen significantly, from one kilogram of CO2e in 2021 to 1.8 kilograms in 2024. Similarly, the carbon cost per million dollars of revenue increased from 26 tonnes of CO2e to 36 tonnes in the same period. This trend shows no signs of slowing. As AI becomes more embedded in our digital lives, the bulk of its energy draw shifts from intentional user queries to background operations, making its environmental impact an unavoidable byproduct of daily digital existence.

The Bottom Line

Meta's aggressive embrace of AI for advertising is a lucrative move for its shareholders but a costly one for the planet. It underscores the urgent need for greater transparency and accountability from tech giants regarding their energy consumption and carbon footprint. As the climate crisis intensifies, the moral valence of every new megawatt-hour of demand becomes critically important. Meta's AI-powered ad tsunami demands we question whether endless digital consumption truly outweighs the environmental cost.