Anthropic, the artificial intelligence company known for developing the Claude language model, has become the first pure AI startup to join Frontier, a prominent carbon removal collective. The move comes as part of a new $915 million funding tranche that Frontier announced, nearly doubling the total amount pledged to the group to $1.8 billion. Since its inception in 2022, Frontier has contracted roughly $700 million across more than 50 carbon removal projects, aiming to eliminate 1.8 million tons of carbon dioxide from the atmosphere.
Frontier was founded by major technology companies including Stripe, Google, and Shopify, with the goal of accelerating the development of carbon removal technologies. These companies, and others that have since joined, face a common dilemma: they want to achieve net-zero emissions within the next decade or two, but certain emissions—such as those from air travel or manufacturing—are difficult to eliminate with current technology. Carbon removal offers a way to offset these residual emissions by funding projects that extract CO2 from the atmosphere and store it permanently.
Anthropic&8217;s participation in Frontier is notable because it marks the first time a company whose primary focus is artificial intelligence has joined the collective. While Google is a founding member, Alphabet&8217;s business spans search, advertising, cloud computing, and hardware, in addition to AI. Anthropic, by contrast, is a relatively young company that has concentrated almost exclusively on building safe and capable AI systems. The decision to invest in carbon removal signals that even AI startups, which have been criticized for their growing energy consumption, are beginning to take climate action seriously.
AI companies have come under increasing scrutiny for their energy use. Training large language models like those developed by Anthropic, OpenAI, and others requires vast amounts of computational power, which in turn consumes significant electricity. While some tech firms have aggressively purchased renewable energy to power their data centers, others have relied on fossil fuels. Anthropic has yet to publish a sustainability report, and the company has described its energy strategy as an &8220;all of the above&8221; approach—a phrase that often implies continued reliance on polluting sources. Joining Frontier may represent a shift in the company&8217;s thinking, as it becomes Anthropic&8217;s first climate-related deal.
Frontier&8217;s model is unique in that it does not simply purchase carbon removal credits for itself. Instead, it acts as an intermediary that vets carbon removal companies, signs contracts with those deemed most promising, and then allocates credits to member companies based on their contributions. This approach allows smaller firms, like Anthropic, to participate without having to build an internal climate team from scratch. It also provides carbon removal startups with long-term, guaranteed revenue streams, which helps them attract additional investment and scale their operations.
Carbon removal credits function similarly to offset credits: companies can subtract them from their carbon footprint, much as profits offset debts on a balance sheet. However, the quality and permanence of different carbon removal methods vary widely. Frontier focuses on technologies that have the potential to store carbon for centuries or millennia, such as direct air capture, enhanced rock weathering, bio-oil injection, ocean alkalinity enhancement, and bioenergy with carbon capture and storage (BECCS). The group has backed a range of projects since 2022, and its criteria have become increasingly stringent over time.
With the latest funding tranche, Frontier is changing its strategy. Instead of making many small bets on a variety of early-stage companies, the collective plans to concentrate on fewer, larger projects that it believes have the best chance of eventually removing one gigaton (one billion metric tons) of CO2 per year. New contracts will typically run for eight to ten years, a shift from the shorter terms of earlier agreements. Frontier also announced that for any new contract it signs, the carbon removal company must demonstrate a path to government subsidy or support. This requirement reflects a broader recognition that private funding alone cannot sustain the industry indefinitely; government involvement will eventually be necessary.
The shift mirrors what appears to be happening at Microsoft, which has been the largest corporate buyer of carbon removal credits. In recent months, Microsoft has also moved toward longer-term, larger-scale contracts. Both companies want to foster a mature carbon removal market, but they are making it clear that they do not intend to underwrite it in perpetuity. The goal is to create an ecosystem where carbon removal becomes a standard business practice, supported by policy frameworks and public investment.
The United Nations Intergovernmental Panel on Climate Change has stated that carbon dioxide removal will be essential for the world to reach net-zero emissions by mid-century. However, few companies or consumers are currently willing to pay the high costs associated with these technologies. Direct air capture, for example, can cost hundreds of dollars per ton of CO2 removed, though costs are expected to fall as the industry scales. Enhanced rock weathering and bio-oil injection are cheaper but face their own challenges, such as land use and verification of permanence.
Anthropic&8217;s involvement may also help raise awareness within the AI industry about the environmental impact of large-scale computing. Several AI companies have recently announced plans to build massive new data centers, and their energy procurement strategies have come under scrutiny. Some have signed power purchase agreements for renewable energy, while others have bought carbon offsets with questionable credibility. By joining Frontier, Anthropic is committing to a more rigorous form of climate action, one that relies on actual removal of carbon rather than simply paying to avoid emissions elsewhere.
The broader implications are significant. If other AI startups follow Anthropic&8217;s lead, the collective purchasing power of the tech industry could accelerate the deployment of carbon removal technologies. Frontier&8217;s new funding tranche brings its total pledges to $1.8 billion, but the organization has already contracted less than half of that. With additional members, it can sign more contracts and drive costs down further. The goal is to create a virtuous cycle: lower costs lead to more demand, which leads to more investment and further cost reductions.
However, challenges remain. Verification of carbon removal is notoriously difficult, and some methods have been criticized for being unproven at scale. Direct air capture plants, for instance, currently capture only a few thousand tons of CO2 per year, far from the millions of tons needed to have a measurable impact. Enhanced rock weathering requires spreading large volumes of crushed rock on farmland, which may disturb ecosystems. Bio-oil injection involves pumping liquefied biomass into geological formations, a process whose long-term stability is still being studied. Frontier and its member companies are betting that with enough funding and scrutiny, these technologies will improve.
Frontier&8217;s new requirement for a government subsidy pathway indicates that the collective expects public sector support to become available. Some countries, including the United States and Canada, have already established tax credits for carbon removal, and the European Union is developing a certification framework. These policies could provide the revenue certainty that carbon removal companies need to build large-scale facilities. In the meantime, Frontier will continue to act as a bridge, channeling corporate dollars into the most promising solutions.
Anthropic&8217;s decision to join Frontier is likely to be seen as a bellwether for the AI industry. As AI companies expand their computing infrastructure, they will face increasing pressure to address their carbon footprints. Those that fail to do so may find themselves at a disadvantage with investors, customers, and regulators. By taking early action, Anthropic is positioning itself as a leader in responsible AI development, even as it continues to push the boundaries of what artificial intelligence can achieve.
The next few years will be critical for both AI and climate tech. If Frontier and its members can demonstrate that carbon removal works at a meaningful scale, they will have created a template for other industries to follow. If not, the world may need to rely even more heavily on emissions reductions, which are already lagging behind what is needed to meet global climate goals. For now, Anthropic&8217;s entry into the Frontier coalition adds a new and influential voice to the effort, and it raises the possibility that other AI companies will soon follow suit.
Source: TechCrunch News