Global climate-tech venture investment reached $26 billion in the first half of 2026, up 55% year-on-year, according to Currence data circulated during the conference. Most of that capital is flowing toward vendors that can sell power to hyperscalers.

Where is the $26 billion actually going?

Nuclear, geothermal, wind and solar firms have signed power purchase agreements with Google, Meta and other operators building AI data centers. Carbon management and low-carbon fuels — both critical for hard-to-abate sectors — saw venture investment plummet in 2026, the same data shows. Those technologies can't easily monetize a megawatt contract.

That imbalance prompted a sharp warning from Simon Stiell, the UN's climate chief. "AI leaders are now on thin ice when it comes to license to operate and sinking deep underwater when it comes to public support," Stiell told the conference. "Tech titans need to start showing why the benefits of AI outweigh its skyrocketing costs — for the many, not just the tiny few."

Why does the data-center buildout matter for R&D managers?

A few years ago, Microsoft, Google and Meta each published aggressive greenhouse-gas reduction targets. All three have since reported emissions increases, primarily tied to data centers supporting AI workloads.

The short-term fix is a wave of new natural-gas generation, which carries a multi-decade asset lifetime and complicates any later decarbonization pathway.

Evelyn Wang, MIT's vice president for energy and climate, told the Associated Press she expects data centers to stop adding to planet-warming emissions within roughly a decade. In a separate panel she pointed to AI's potential to speed up progress in the search for new catalysts — a concrete workflow gain that R&D leaders in chemicals and materials can already begin to plan around.

Has the 1.5°C window already closed?

A UN Environment Programme report released this month concluded that the world has nearly passed the threshold for holding warming under 1.5°C above preindustrial levels. The report recommends aggressive emissions cuts paired with large-scale carbon removal.

That portfolio requires sustained R&D spend in direct air capture, geological sequestration and engineered sinks — not just clean-power capacity.

UN Secretary-General António Guterres framed the AI question as binary at the assembly's opening. "The climate crisis fuels instability and displacement," Guterres said. "Artificial intelligence could help solve all these challenges, or it could make them worse."

What should R&D leaders track next?

Three near-term signals matter for portfolio and capital planning:

• Whether hyperscalers publish updated emissions trajectories with interim milestones rather than renewable-procurement totals alone
• The next quarterly Currence readout on climate-tech flows, which will reveal whether 2026's carbon-management slump continues or reverses
• Local permitting outcomes for proposed gas plants paired with data centers, since each approval adds decades of fossil lock-in to regional grids

Public opposition is already visible near active sites, with residents citing noise and air-quality complaints. Combined with Stiell's warning about AI leaders' "license to operate," the signals suggest the political ceiling on a gas-heavy buildout is lower than today's deal pipeline implies.

The question researchers and R&D managers will answer through 2027 is whether the $26 billion flowing into climate tech is reshaping the energy mix fast enough to offset the emissions embedded in the infrastructure now being financed — or whether the gap will need to be closed by the carbon-removal technologies that just lost funding momentum.