Technology

Meta Carbon Emissions Soar 45% as AI Data Centers Grow

Meta Carbon Emissions Soar 45% as AI Data Centers Grow

Meta has announced it is re-evaluating its climate goals, including its 2030 net-zero emissions target, as the rapid expansion of artificial intelligence drives up energy consumption and carbon pollution.

Climate Goals and AI Expansion

Meta previously promised to become carbon-neutral and water-positive by 2030, meaning it would offset all created carbon dioxide and return more water to local watersheds than it consumed. However, the company is adjusting those plans to accommodate the immense power needs of artificial intelligence and the data centers required to operate it.

In Meta’s sustainability report released on Thursday, the company’s data center and sustainability leaders stated that they are re-evaluating timelines, sequencing, and interim milestones, specifically mentioning the 2030 net-zero emissions goal and near-term intensity targets. A Meta spokesperson told CNET via email that while the company remains committed to net zero, it is reassessing the 2030 timeline and will continue investing in emissions reductions.

Meta joins other major technology companies such as Microsoft, Amazon, and Google in retreating from previous climate targets to construct hyperscale data centers for AI workloads. Microsoft President Brad Smith previously noted in 2023 that his company’s net-zero target was set before the artificial intelligence boom.

Rising Emissions and Energy Consumption

According to the report, Meta recorded net emissions of 11.8 million metric tons of carbon dioxide in 2025, marking a 45% increase compared to the previous year. Greenhouse gases are the primary drivers of the climate crisis, fueling extreme weather patterns and natural disasters.

The infrastructure required to support these systems is immense, as seen with Meta’s Hyperion data center in Louisiana. Announced by CEO Mark Zuckerberg in 2025, the facility will rival the size of Manhattan when it becomes operational in 2030, generating two gigawatts of AI capacity with potential for future scaling while placing heavy demands on the local energy grid.

Meta outlines various mitigation strategies in its report, such as enhancing water-use efficiency in data centers and reducing embodied carbon in new designs by roughly 77% compared to original models.

Despite these measures, the company’s electricity consumption surged by 35% year over year as new data centers began operating. Meta stated that demand for AI infrastructure has accelerated at an unprecedented pace.

You May Also Like

Leave a Comment

Your email address will not be published. Required fields are marked *