Anthropic, the artificial intelligence company behind the Claude model family, has made its first major climate move by joining Frontier, the carbon removal collective founded by Stripe, Google, and Shopify. The startup contributed to a new $915 million tranche of funding, making it the first pure AI company to join the ranks of Frontier's members. The announcement, made on June 17, 2026, nearly doubles the total pledges to Frontier, bringing the collective's committed funding to $1.8 billion.
Frontier was launched in 2022 to accelerate the development of carbon removal technologies by guaranteeing future demand. Since its inception, the group has contracted nearly $700 million across more than 50 projects to remove 1.8 million metric tons of carbon dioxide from the atmosphere. Companies that pledge money to Frontier typically use the carbon removal credits generated by these projects to offset their unavoidable emissions and reduce their publicly reported carbon footprints.
Anthropic's decision to join Frontier is notable not only because it is the first AI startup to do so but also because it comes at a time when the AI industry has been on an unprecedented energy buying spree. AI companies have been securing massive amounts of power to run their data centers, often through deals that include fossil fuel-based electricity. Anthropic itself has not produced a sustainability report and has previously stated that it favors an "all of the above" approach to energy — a stance that often translates into significant purchases of polluting power. The move to join Frontier may signal changing attitudes within the company or a recognition that the AI sector needs to address its growing climate footprint.
Frontier was created to solve a chicken-and-egg problem: many tech companies want to reach net-zero emissions by 2030 or 2040, but they have residual emissions from activities like air travel and supply chains that cannot be eliminated with current technology. At the same time, carbon removal was — and still is — a nascent industry dominated by small startups that cannot yet remove carbon at the scale needed. Frontier vets carbon removal companies and signs multiyear contracts for credits, providing the revenue certainty that allows these startups to scale.
How carbon removal credits work
Carbon removal credits function like a financial offset. Companies can purchase credits that represent a verified ton of CO2 removed from the atmosphere and stored permanently. These credits are then subtracted from the company's carbon footprint, similar to how profits counterbalance debts on a balance sheet. Unlike carbon offsets that merely avoid emissions (such as protecting a forest that might otherwise be cut down), removal credits physically extract CO2 that is already in the atmosphere. Frontier focuses exclusively on removal credits and applies rigorous vetting to ensure the carbon is stored for at least hundreds of years.
Anthropic's climate journey
Anthropic's entry into Frontier marks its first climate-related deal. The company, which has rapidly grown in valuation and influence since its founding in 2021, has been largely silent on environmental issues. It has no sustainability report published as of early 2026, and its public statements on energy procurement have focused on reliability and cost rather than carbon intensity. However, the company's leadership has expressed awareness that AI's energy demand is soaring. By one estimate, AI data centers could consume as much electricity as a small country by 2030. Joining Frontier could be a signal that Anthropic intends to take responsibility for its share of that impact, even as it continues to grow.
The broader AI industry has faced increasing scrutiny over its carbon footprint. Training large language models requires massive amounts of electricity, and the inference phase — when models are deployed to answer queries — can be even more energy-intensive. Companies like OpenAI, Microsoft, and Google have all made varying degrees of climate commitments, but critics argue that the pace of AI expansion is outstripping the buildout of clean energy. Anthropic's move could put pressure on other AI startups to follow suit.
Frontier's new direction
With the new $915 million tranche, Frontier also announced a shift in its investment strategy. The organization said it will fund fewer projects in the future, focusing instead on those with the greatest potential to scale to a gigaton (1 billion metric tons) of CO2 removal per year or more. New contracts will typically run for eight to ten years, providing longer-term certainty for developers. This shift reflects the maturation of the carbon removal industry: after an initial phase of placing many small bets to stimulate innovation, Frontier is now concentrating its capital on the most promising technologies.
Frontier has supported a diverse array of carbon removal methods since its launch, including direct air capture (DAC), enhanced rock weathering, bio-oil injection, ocean alkalinity enhancement, and bioenergy with carbon capture and storage (BECCS). Each technology has different cost profiles, storage durations, and co-benefits. For example, DAC facilities can be located anywhere but are energy-intensive, while enhanced weathering uses crushed silicate rocks spread on farmland to accelerate natural chemical reactions that absorb CO2. Frontier's new strategy does not rule out any specific approach but will emphasize robust measurement, reporting, and verification (MRV) standards.
The role of government
Frontier's leadership has made it clear that they do not intend to underwrite the carbon removal industry indefinitely. For any new contract signed, the carbon removal company must "show a path to government subsidy/support," a Frontier spokesperson told TechCrunch. This mirrors the trajectory of many clean technologies, such as solar and wind power, which initially relied on government incentives and guarantees before becoming cost-competitive. The UN Intergovernmental Panel on Climate Change (IPCC) has repeatedly stated that carbon dioxide removal will be necessary to achieve net-zero emissions by mid-century, but voluntary corporate purchases alone cannot meet the required scale.
The transition from private to public funding is already visible in the carbon removal space. The U.S. Department of Energy has announced billions of dollars in funding for DAC hubs under the Infrastructure Investment and Jobs Act. Other governments, including the United Kingdom, Canada, and the European Union, are developing similar programs. Frontier's demand for a government pathway suggests that the collective sees public support as essential for the industry's long-term viability. The group said it will contract as far out as 2040, but it did not specify what happens after that. The implicit hope is that governments will have started to take the reins by then.
Comparison with Microsoft's approach
Microsoft has been the largest single buyer of carbon removal credits, purchasing over 5 million tons of credits through various agreements. The company has also shifted its strategy in recent years, moving from a broad portfolio of small projects to fewer, larger contracts with companies like Heirloom Carbon and Climeworks. Microsoft's approach has influenced the wider market, and Frontier's decision to concentrate its investments mirrors this trend. However, Microsoft has been able to leverage its size to negotiate favorable terms, while Frontier's collective model allows smaller companies like Anthropic to pool resources and share the vetting burden.
The bigger picture
The timing of Anthropic's entry into Frontier is significant. The AI industry is under growing pressure to decarbonize its operations, and carbon removal credits offer a way to address residual emissions that cannot be eliminated through efficiency or clean energy purchases. However, critics argue that credits should not be used as a substitute for direct emissions reductions. Frontier's rigorous vetting process is designed to ensure that the credits represent real, durable removals. Still, the broader climate community remains divided on whether carbon removal should be a primary tool or a last resort.
Anthropic's membership may also embolden other AI companies to join frontier collectives or develop their own carbon removal strategies. With total pledges now at $1.8 billion, Frontier has the resources to fund the next wave of carbon removal startups. The collective has already contracted for 1.8 million tons of removal, but that is a tiny fraction of the 10 billion or more tons that experts say will be needed annually by mid-century. The gap between ambition and reality remains vast, but each new member and each new dollar brings the industry a step closer to viability.
For now, Anthropic is charting its own course in climate action. The company has yet to release a comprehensive sustainability report, and its energy procurement strategy remains opaque. But joining Frontier is a tangible commitment that could be followed by other steps. As the AI sector continues to expand its energy footprint, the industry's willingness to invest in carbon removal will be closely watched by policymakers, investors, and environmental advocates alike. The success of Frontier's new strategy — fewer projects, longer contracts, and a push for government support — will determine whether carbon removal can scale fast enough to make a meaningful difference in the fight against climate change.
Source:TechCrunch News