Anthropic has become the first pure AI startup to join Frontier, the carbon removal collective, contributing to a new $915 million funding tranche that nearly doubles the coalition's total pledges to $1.8 billion. The announcement, made on June 17, 2026, marks a significant milestone for both Anthropic and the carbon removal industry, which has historically relied on tech giants and financial firms for support.
Founded in 2022 by payments company Stripe, along with Alphabet (Google), Shopify, and McKinsey Sustainability, Frontier was designed to accelerate the development of carbon removal technologies by guaranteeing future demand. The collective acts as an advance market commitment, vetting early-stage carbon removal companies and signing contracts to purchase their credits once they become operational. To date, Frontier has contracted nearly $700 million across more than 50 projects to remove 1.8 million tons of carbon dioxide from the atmosphere. The new $915 million injection will allow Frontier to expand its portfolio and shift its strategy toward fewer, larger projects with a higher chance of achieving gigaton-scale removal—that is, removing one billion metric tons of CO₂ annually.
Anthropic’s participation is particularly notable because it represents the first time a company whose primary business is artificial intelligence has joined the coalition. While Google, a founding member, is heavily involved in AI, its parent company Alphabet has a diversified revenue stream. Anthropic, by contrast, is solely focused on building large language models and AI safety. The startup has been on an energy buying spree to power its data centers, and much of that energy has come from conventional sources, raising questions about its carbon footprint. Until now, Anthropic had not engaged in any climate-related deals, nor had it published a sustainability report. The company had expressed a preference for an “all of the above” energy strategy, which often includes natural gas and other fossil fuels alongside renewables. By joining Frontier, Anthropic is signaling a potential shift in its approach, though it remains to be seen how the company will balance its growing energy needs with its climate obligations.
The timing of Anthropic’s move is crucial. AI companies have become some of the largest consumers of electricity, with data centers for training and inference consuming vast amounts of power. Many of these companies have faced criticism for not doing enough to procure clean energy or offset emissions. Anthropic’s membership in Frontier could pressure other AI firms, such as OpenAI, Cohere, or Mistral, to follow suit. However, the cost of carbon removal credits remains high—often hundreds of dollars per ton—and the technology is still in its infancy. Critics argue that carbon credits allow companies to delay real emissions reductions, but proponents maintain that removal is necessary for the emissions that are impossible to abate in the near term, such as those from air travel or manufacturing.
Frontier’s new funding round comes with increased scrutiny. The collective announced that it will fund fewer projects moving forward, focusing on those it believes have the best chance of reaching gigaton-scale removal. Contracts will run for eight to ten years, longer than previous agreements. Frontier also stated that any new contract must include a clear path to government subsidy or support, signaling that the coalition does not intend to underwrite the carbon removal market indefinitely. This mirrors a trend seen at Microsoft, which has been the largest corporate buyer of carbon removal credits. Microsoft has also begun demanding that projects demonstrate viability without assuming permanent corporate backing.
The carbon removal industry encompasses a variety of technologies. Frontier has historically backed direct air capture (DAC), which uses chemical processes to pull CO₂ from the ambient air; enhanced rock weathering, which accelerates natural mineral reactions that absorb CO₂; bio-oil injection, which stores carbon in liquid form underground; ocean antacids, which reduce ocean acidification while sequestering carbon; and bioenergy with carbon capture and storage (BECCS), which burns biomass and captures the resulting CO₂. Each technology has its own cost structure, scalability challenges, and environmental trade-offs. The shift toward larger, more vetted projects suggests Frontier is betting on a handful of approaches that can achieve dramatic scale, likely with the help of future government mandates or subsidies.
The role of governments in carbon removal has been a recurring theme. The UN Intergovernmental Panel on Climate Change (IPCC) has stated that carbon dioxide removal will be necessary to achieve net-zero emissions by mid-century. Yet few companies or consumers are willing to pay the current high cost of removal. Frontier, by providing early funding, hopes to drive costs down through learning curves and economies of scale. The collective has contracted as far out as 2040, after which it expects governments to take over as primary customers. If that does not happen, the consequences could be severe: the rate of climate change is accelerating, and without large-scale removal, meeting global temperature targets will become nearly impossible.
Anthropic’s decision to join Frontier also raises questions about the broader AI industry’s environmental impact. The energy consumption of AI models has been a subject of intense debate. Training a single large model can emit as much carbon as five cars over their lifetimes, and inference—the process of running models in production—demands even more energy as usage scales. Some AI companies have invested in renewable energy certificates or bought carbon offsets, but these measures are often criticized for lacking additionality or permanence. By joining a rigorous coalition like Frontier, Anthropic is choosing a more direct approach to addressing its residual emissions, though the credits it purchases will not negate the need for actual emissions reductions.
The broader context of corporate climate action is also relevant. Many of the world’s largest companies have pledged to achieve net-zero emissions by 2030 or 2040, but progress has been uneven. Carbon removal credits are one tool in the toolbox, but environmental advocates warn against over-reliance. Frontier’s vetting process is designed to ensure that credits represent real, durable carbon removal, but even the best credits cannot substitute for deep cuts in fossil fuel use. Anthropic, by joining Frontier, is acknowledging that some emissions are unavoidable for now, but the company must also pursue energy efficiency, renewable procurement, and operational changes to reduce its overall footprint.
In summary, Anthropic’s entry into Frontier marks a pivotal moment for both AI and climate technology. It signals that AI companies, despite their massive energy consumption, are beginning to engage seriously with the carbon removal market. Whether this trend will continue depends on the cost trajectory of removal technologies, the regulatory environment, and the willingness of other AI startups to follow Anthropic’s lead. For now, Frontier has gained a new member that brings both financial resources and symbolic momentum to the fight against climate change.
Source: TechCrunch News