nonAI

When everyone runs away, Meta’s former CTO stays. While the majority of Silicon Valley investors have abandoned the C thesisfile Tech —tired of promises that do not turn into real business—, Mike Schroepfer just announced that it has raised 250 million dollars to do exactly the opposite. Its background, Gigascale Capitalhas closed its first round with institutional investors to back founders who, in their own words, are “rebuilding the brick-and-mortar economy.” The news comes at a time when the climate technology sector has a difficult reputation. As explained TechCrunch, “Conventional wisdom” has been soured by the “Climate Tech” label. Schroepfer, known in the industry simply as Schrepis challenging the market consensus. Or as the same media describes it: “Zigging when most are zagging” (zigzagging when everyone goes in the other direction). Does this bet make sense? First of all, we must understand the underlying problem: gas turbines, the most conventional electricity generation system that exists, currently have a waiting list that extends until the early 2030s. It is not that there is a lack of green technology, it is that there is simply a lack of energy. And companies trying to connect to the electrical grid are finding it increasingly difficult. The person responsible. And the question is, who has accelerated that demand to this point? Artificial intelligence. The sector has undergone a structural change in recent years, driven precisely by the energy demands of AI. Data centers consume huge amounts of electricity and networks cannot cope. Faced with this situation, many companies are trying to generate their own electricity. As Schroepfer himself notedthe “bring your own energy” model (Bring-Your-Own-Power) will become a decisive competitive advantage in intensive industries. But there is no easy path there either: even traditional turbines have a waiting list. As Pulse 2.0 detailsaccelerated electrification, industrial relocation, AI deployment and increasingly extreme climate events are simultaneously putting pressure on physical infrastructure that has been aging for decades. The business of scarcity. The company, Gigascale, was founded in 2023 by Schroepfer along with Victoria Beasley and Evaline Tsai. The fund emerges from a process that the former Meta executive describes as a systematic study of the climate sector during the pandemic. In three years they have built a portfolio of more than 25 companies in areas ranging from clean energy and grid infrastructure to critical minerals, advanced manufacturing and what they call “physical AI”: applications of artificial intelligence to design, manufacture and deploy real-world systems. Schroepfer’s investment logic does not pivot on environmental virtue, but on competitiveness. Their argument is the following: solar went from producing 40 gigawatts a year to 600 in a decade because it became cheaper. “The companies we support win because they are cheaper, faster and more reliable. This is how adoption scales. Climate impact is the result of systems that work better,” declared in a statement. When the waiting list is the opportunity. The fund’s portfolio already has specific names that illustrate this philosophy: New energy generation: Commonwealth Fusion Systems and Xcimer Energy (which achieved the first flash of its commercial laser system in late 2025) are working to make nuclear fusion a reality. For its part, Radiant is moving toward one of the first commercial deployments of nuclear microreactors in the United States. Infrastructure for AI: Arbor Energy has signed an agreement with GridMarket to supply up to 5 gigawatts of clean, zero-emissions energy to data centers. In parallel, Fractile announced a $136 million expansion to manufacture AI processors specifically designed to reduce electricity consumption. Circular and industrial economy: Heron Power, founded by Drew Baglino – former Tesla vice president for propulsion and energy division – develops industrial power electronics. In addition, companies like Dioxycle have signed multi-year agreements with giants like L’Oréal to convert captured CO₂ emissions into ethylene to make packaging. There is an underlying irony. The world has been debating for years how to decarbonize for environmental reasons. And it turns out that the catalyst that is making the transformation of the energy system urgent and inevitable is not any climate summit: it is Artificial Intelligence. As investors flee the label Climate Tech Considering it too ideological or unprofitable, the demand for energy is so brutal that not even the most conventional gas turbines can cope. The opportunity exists precisely because the problem is real. And Schroepfer, who comes from building the systems that consume that energy, is very clear about it. Image | Unsplash Xataka | From “tokenmaxxing” we have moved on to “tokenwasting”: the level of waste in AI is reaching unprecedented levels