NEW YORK — Artificial intelligence has taken over the conversation at this year's Climate Week, turning a gathering usually centered on emissions targets, clean-energy finance and policy pledges into a broader reckoning over the future of both climate action and the technologies now reshaping it.
With world leaders in Manhattan for the UN General Assembly and climate investors, policymakers, advocates and journalists moving between panels, private talks and high-profile dinners, the debate has grown louder than usual. This year, the unavoidable topic is AI.
The tension is not hard to understand. In climate circles, AI is being discussed as both a tool and a threat: a technology that could accelerate scientific discovery, improve energy systems and help optimize industrial processes, while also driving a surge in electricity demand that is already reshaping power markets and emissions trajectories.
Everyone seems to agree that AI matters. The disagreement is over how its influence will unfold.
UN Secretary-General António Guterres put that duality at the center of his remarks on the first day of the General Assembly. "The climate crisis fuels instability and displacement," Guterres said. "Artificial intelligence could help solve all these challenges, or it could make them worse."
That warning lands at a moment when climate ambition is under obvious strain. A recent report from the UN Environment Program said the world has nearly passed the point where it could keep warming to less than 1.5 degrees Celsius above preindustrial levels. With that threshold slipping out of reach, the report argues that the world must not only move rapidly to cut greenhouse-gas emissions, but also scale carbon removal to help absorb emissions already in the atmosphere.
The question echoing through Climate Week is whether AI can help with any of that.
On one hand, the technology's energy appetite is forcing a long-overdue focus on electricity supply, grid reliability and the buildout of new generation. That has translated into more attention — and more capital — for energy technologies, including some that are low- or zero-emissions. Startups across nuclear, geothermal, wind and solar have signed deals with companies such as Google and Meta as the tech giants look to power expanding data centers.
The money is following the demand. Global climate-tech venture investment reached $26 billion in the first half of 2026, according to data from Currence, a finance tracker for the industry. That is 55% higher than a year earlier, and products and services tied to data centers are taking a large share of that investment.
But the boom is not lifting all parts of the climate-tech ecosystem equally. As a Semafor report on the data noted, carbon management and low-carbon fuels have seen venture investment fall sharply this year. Those sectors remain central to climate strategy, but they do not benefit as directly from the data-center buildout as power-generation and grid-related businesses do.
The result is a climate-finance market increasingly shaped by AI's infrastructure needs rather than by the full range of decarbonization priorities. That shift is already visible in corporate emissions data. Microsoft, Google and Meta all once set ambitious goals to reduce greenhouse-gas emissions, but each has since seen emissions rise, largely because of the data centers required to support AI.
For some at Climate Week, that is evidence that the industry is moving in the wrong direction, trading long-term climate commitments for near-term digital expansion. For others, it is a sign that AI is forcing the kind of investment in energy systems that climate advocates have sought for years — even if the catalyst is not climate policy itself.
There is also a more optimistic camp arguing that AI could speed progress in areas that have long been bottlenecked by time, cost and complexity. One example often raised is materials discovery, including the search for new catalysts that could improve industrial processes or make clean-energy technologies more efficient. Evelyn Wang, MIT's vice president of e... [truncated due to source material ending]
What makes this year's Climate Week different is not simply that AI is present, but that it is now unavoidable in nearly every climate conversation. It is influencing where capital flows, how utilities plan for load growth, how technology companies justify their infrastructure spending and how climate advocates think about the next phase of decarbonization.
That makes the debate especially consequential. If AI helps unlock breakthroughs in energy, materials science and carbon removal, it could become one of the most important accelerants in the climate fight. If, instead, it deepens dependence on fossil-fueled power and diverts attention from harder-to-finance climate solutions, it could make an already difficult transition even more complicated.
For now, Climate Week is reflecting that uncertainty in real time. The conversations are animated, the stakes are high, and the answer to AI's climate role remains unsettled. But in a year when the world is confronting the reality that the 1.5°C target is slipping away, the technology's rise has made one thing clear: the climate debate is no longer only about how to cut emissions. It is also about which technologies will define the next phase of the energy transition — and at what cost.