Altman Dismisses AI Water Concerns, Sparks Debate on Energy and Human Value

OpenAI CEO Sam Altman has labeled widespread concerns about AI's water consumption as "fake," while acknowledging the broader energy challenge. Speaking at the India AI Impact summit, Altman refuted viral claims that ChatGPT uses gallons of water per query, calling them "completely untrue" and "totally insane." He argued that while data center cooling has traditionally required water, newer facilities often operate without it.

However, Altman conceded that the sector's total energy draw is a legitimate issue, stating the world must accelerate its shift to nuclear, wind, and solar power to meet growing AI demand. He also drew a controversial parallel between AI and human energy use. "It takes a lot of energy to train a human," Altman said, referencing the years of sustenance and education required. He suggested that on a per-query basis, after initial training, AI may already match human energy efficiency.

This comparison drew immediate criticism. Sridhar Vembu, co-founder of Zoho Corporation, who was present at the summit, later stated on social media, "I do not want to see a world where we equate a piece of technology to a human being."

The debate unfolds against a backdrop of rapid data center expansion. An International Monetary Fund report noted that global data center electricity use in 2023 rivaled that of major European nations. This growth has led to local pushback, such as in San Marcos, Texas, where a $1.5 billion data center project was recently rejected over grid strain and cost concerns. Altman and other tech leaders maintain that meeting this demand will require a major, diversified increase in energy production, particularly from non-carbon sources.

Source: CNBC