Technology news and discussion on TalkFuse.
Even the ‘good guys’ of AI are locking in fossil fuels
Anthropic, which has built a reputation for prioritising AI safety, is facing criticism over the climate impact of the energy used to power its data centres. Greenpeace says a reported $3.4 billion agreement for gas-fired engines would commit the company to fossil fuels as electricity demand from AI grows, raising questions about whether its environmental actions match its responsible image.
The article says the September deal with Texas-based OnePWR Solutions is the largest known single purchase of its kind by an AI lab. Anthropic’s IPO filings describe climate change mainly as a business risk, including higher cooling costs, operational disruption from extreme weather and possible damage to its brand; they also note that sustainability measures and disclosure rules could be costly. The article contrasts this with Anthropic’s public focus on AI alignment and safety.
In short
- Anthropic faces criticism over a reported $3.4 billion gas power deal.
- Its filings discuss climate change largely as a business risk.
- The deal raises questions about its responsible-AI reputation.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Anthropic could argue that rapidly growing data-centre demand requires dependable power, and that gas-fired engines can provide electricity when renewable supply is unavailable or the grid is constrained. Securing capacity may help it keep services reliable while buying time for cleaner energy and infrastructure to scale; its filings also acknowledge climate-related risks and the costs of mitigation and disclosure.
The case against
Critics could argue that a large, long-term commitment to gas risks locking in emissions and making fossil-fuel infrastructure harder to retire just as AI’s electricity demand is rising. A company that presents itself as responsible should account for the climate effects of its operations, not only the financial risks to its business, and should prioritise cleaner power and transparent environmental commitments.
Explained simply
Anthropic is an artificial intelligence company known for developing Claude and for its public emphasis on making AI systems safe. AI services run on data centres, which use large amounts of electricity to power and cool computer equipment.
Gas-fired engines generate electricity by burning natural gas, a fossil fuel that releases greenhouse gases. A reported agreement between Anthropic and Texas-based OnePWR Solutions has drawn attention to how quickly the company’s energy needs may grow as it expands its AI services.
The debate concerns how companies balance reliable power for data centres with the climate effects of the energy they use. Anthropic’s IPO filings discuss climate change mainly in terms of potential business costs and risks, while its public profile centres on AI safety.