Beyond the AI Bubble: The Technologies That Deserve the Hype

Nuclear Fusion, Asteroid Mining, and all the Innovation you're not hearing about.

Whether the AI bubble is about to burst or not, you’re just wondering, “Is there anything in life more than AI?”

As a Data and Computer Scientist, I remember that before the release of the first ChatGPT in 2022, the startup ecosystem was diverse and rich. You walked into any accelerator demo day, and you’d encounter in one corner a crypto team pitching decentralized exchanges, while next to them, biotech founders were showing off precision medicine frameworks built on personal genomic data. Fintech startups were building banking infrastructure for the unbanked population in Africa. Hardware accelerators for crypto. Home automation startups that weren’t just Alexa-enabled.

There was variety, and what was common to everyone was the need for computer scientists. At that time, a computer scientist could transition from biotech to fintech, from hardware optimization to visually-appealing dashboards. It was an enriching experience.

My first job was actually applying statistical models to treat patients with rare diseases. It was an exciting period. My MSc thesis was about applying a graph-based algorithm to solve drug repurposing during COVID-19. While working, I took bioinformatics classes, and I could easily follow technical speeches in biology.

Now, GenAI has monopolized capital and talent. I feel like I am pulled into a tunnel, and I have to intentionally seek news from other channels. Despite the incredible advancements, I remain unimpressed by the quality of AI-generated content: I have yet to find a single application that justifies all this capital waste and marketing.

However, skepticism apart, which many people may disagree with, we can all agree that one of the first rules in business is to never put all your eggs in one basket.

So, if I were an investor, where do I invest that is not AI-related? We have other problems to solve and tackle that, in my opinion, are more urgent, like energy, making our environment cleaner, reducing pollution, and exploring new frontiers of computing beyond the Von Neumann machine.

Let’s dive in.

Zero Water Cooling Data Centers

We all know that Generative AI has massive environmental costs that extend beyond just the electricity bill, like carbon emissions, water consumption, and grid strain. A single ChatGPT query uses 5x more electricity than a web search. Manufacturing GPUs has a higher carbon footprint than CPUs due to the complex fabrication process, which involves dirty mining and higher toxic waste. Data centers require up to 1.5 million litres of water per day per hyperscale data centre, which may disrupt local ecosystems and leave vulnerable populations without access to clean water. Full paper here: https://mit-genai.pubpub.org/pub/8ulgrckc/release/2

Do we have startups that promise to solve this HUGE environmental issue?

Apparently, yes. Multiple startups are tackling this: companies like Asperitas, Iceotope, and Submer are developing immersion and liquid cooling systems that dramatically reduce or eliminate water consumption. Microsoft announced in 2024 that all new data centers will use zero-water closed-loop cooling systems, potentially saving over 125 million liters per facility annually.

We’re using innovation to clear after innovation. It’s in everybody's interest to find solutions to reduce water drainage, so I’m confident that zero-water closed-loop systems will become the new standard.

Nuclear fusion

One thing is sure about our civilization: we need massive and clean energy sources. While solar and wind are essential, they’re incremental.

What about nuclear fusion, which promises to solve the energy problem forever?

The fusion startup ecosystem has raised over $7 billion total, with companies like CFS (Commonwealth Fusion Systems) ($2B+ raised), Helion ($1B+), and General Fusion actively building demonstration reactors. Helion has an aggressive timeline to produce electricity by 2028, with Microsoft as its first customer. CFS promises that in 2027, SPARC will become the world’s first commercially relevant fusion energy machine to produce more energy from fusion than it needs to power the process. There are more than 40 nuclear fusion companies globally working on various approaches like tokamaks, stellarators, and field-reversed configurations. https://climateinsider.com/2024/10/04/nuclear-energy-startups/

Lab-grown meat

This sector is very important to me because I’m against animal cruelty and because factories are the most significant contributors to greenhouse gas emissions, water contamination, and air pollution.

SuperMeat achieved cost parity with premium pasture-raised chicken at $11.79/pound in 2025. Believer Meats, in 2021, announced that it could produce almost a pound of cultivated chicken for $7.70, with a 110-gram chicken breast costing $1.70 to manufacture. Eat Just’s cultivated chicken is commercially available in Singapore and has become the first company, alongside Upside Foods, to have its cultivated chicken approved by the U.S. Department of Agriculture (USDA), marking the first time any cultivated meat has received approval in the U.S.

Companies like Aleph Farms, Mosa Meat, and dozens of others are working on beef, pork, and seafood. The sector is real and scaling.

Cultured Meat: https://www.labiotech.eu/best-biotech/cultured-meat-companies/

Clean Startups: Plastic Recycling

The problem with traditional plastic recycling technology is that it degrades the quality of the material until it becomes worthless, thereby only deferring its trip to the landfill or the ocean. These startups may solve the problem by breaking down plastic and rebuilding it to virgin quality, thereby creating an actual circular economy. If they succeed, goodbye to oil-based production entirely!

We have PureCycle Technologies ($844M raised), which recycles polypropylene plastic, removing all odor, color, and contaminants.

Loop Industries ($140M) depolymerizes waste PET plastic back into monomers to remake virgin-quality plastic.

Samsara Eco ($107M) utilizes enzymes to break down plastic into its core molecules.

DePoly (Swiss startup) chemically converts mixed post-consumer PET plastic, textiles, and fibers into virgin-grade purified terephthalic acid (PTA) and monoethylene glycol (MEG), the raw materials used to produce new PET without relying on oil.

More here : Recycling startups

Waste Heat Recovery

Waste heat recovery is the process of capturing and reusing excess heat generated by industrial or other processes that would otherwise be released into the environment. It may not be the sexiest innovations, but the ROI is there.

Orcan Energy (Germany, €28.5M) uses Organic Rankine Cycle technology to convert low-temperature waste heat (the largest share of waste) into clean electricity. Sierra Energy heats trash to 4,000 degrees and harvests the gas to make plastics and fuel

The waste heat recovery market is projected to reach $114.7 billion by 2028. CEMEX partnered with Orcan Energy for waste heat recovery at its cement plant in Brandenburg. BASF and MAN Energy Solutions are building an industrial-scale heat pump at BASF’s Ludwigshafen site.

https://aster-fab.com/industrial-waste-heat-recovery-technologies-and-applications/

Asteroid mining

We’re running out of iridium and platinum within a decade. Metallic asteroids contain concentrated deposits of these critical metals, offering access to resources that we cannot reach on Earth without drilling thousands of kilometers into our own core. Asteroid mining may not be the wisest short-term investment, but within a decade, it will become a necessity.

AstroForge launched the first commercial deep space mission targeting an asteroid in February 2025 (although it failed), and plans to land on a metallic near-Earth asteroid in 2026. The company has raised $55 million. Other companies, such as TransAstra and Asteroid Mining Corporation, are developing optical mining technology and robotic explorers [Nasdaq].

Quantum Computing

Quantum computers have the potential to revolutionize science. For example, a large-scale quantum computer could break some widely used public-key cryptographic schemes and aid physicists in performing physical simulations.

Quantum startups raised $1.9 billion in 2024, a 138% jump from 2023. Major rounds include QuEra Computing’s $230 million and Quantum Machines’ $170 million Crunchbase News. Alice & Bob raised $104 million in January 2025 for fault-tolerant quantum computers, and PsiQuantum is building a million-qubit machine with substantial government backing TechCrunch.

So, is there more to Chatbots? Hell, yeah.