A wave of deep-tech startups has raised over $400 million to reclaim rare-earth elements from electronic waste. By turning broken technology into a sustainable resource, these companies are bypassing traditional mining to extract critical elements right from our discarded gadgets.
Forget plant-based burgers and laboratory-grown steak. A fiercely competitive cohort of deep-tech biotech startups is shaking up the venture capital world by commercializing an entirely new category of nutrition: protein brewed directly from industrial carbon dioxide emissions. By leveraging ancient, gas-fermenting microbes originally studied by NASA, these companies are transforming factory pollution into high-purity protein effectively decoupling the future of global food security from the limitations of land, water, and shifting climate patterns.
In December 2025, Meta Platforms announced the acquisition of Manus, a Singapore-based AI startup whose general-purpose AI agents could function as digital employees, independently carrying out tasks such as research and automation with minimal human input. The price was approximately $2 billion. The deal closed. Manus employees moved into Meta offices in Singapore. Capital was transferred. The startup's executives joined the American firm's rapidly expanding AI team. Investors including Tencent Holdings, ZhenFund, and HSG received their proceeds. The acquisition was, by all appearances, complete.
William Bryk has a prediction that sounds absurd until you realize the data already backs it up. Sometime this year, he says, AI agents will conduct more searches on the internet than humans. Not someday. Not in a decade. This year. “There are billions of humans doing searches,” he told Bloomberg this week. “There are going to be trillions of agents, like, very soon.”
Sometime in the next several weeks, a Swedish private equity firm with €269 billion in assets will begin raising a fund that represents something Europe has never attempted before: a concerted, state-backed effort to stop the continent's most valuable technology companies from being systematically acquired by American and Chinese buyers before they reach scale. The Scaleup Europe Fund, a €5 billion vehicle that will target late-stage companies in artificial intelligence, quantum computing, space technology, clean energy, and biotechnology, is the most ambitious government intervention in European venture capital since the creation of the European Investment Fund in 1994. It is also an admission of failure.
Sometime this week, the U.S. Securities and Exchange Commission is expected to publish a document that will do something no American financial regulator has ever done. It will create a legal framework for two parallel stock markets to trade the same shares of Apple, Tesla, and Nvidia—one running on the century-old infrastructure of the New York Stock Exchange, the other on crypto rails that settle instantly, operate 24 hours a day, and require no permission from the companies whose shares they track. The document is called an "innovation exemption," and it is the most significant regulatory experiment in the structure of American equity markets since the creation of the SEC itself in 1934.
For as long as quantum computers have existed, they have been haunted by a single, unforgiving problem: noise. The same quantum properties that make qubits so powerful—their ability to exist in multiple states simultaneously, to entangle with one another across distances, to perform calculations that would take classical computers longer than the age of the universe—also make them exquisitely fragile. A stray photon. A fluctuation in temperature. A whisper of electromagnetic interference from a nearby power line. Any of these can cause a qubit to lose its quantum coherence, introducing errors that cascade through a calculation and render the output meaningless.
India’s edtech valuation reset increasingly reflects broader shifts as startup ecosystems move from hypergrowth narratives toward sustainability and business durability.
In the autumn of 2020, a small Finnish game studio released a mobile puzzle title that should not have worked. The game involved no shooting, no racing, no combat of any kind. Its central mechanic was merging—dragging one garden tool onto another to create a slightly better garden tool, then using that tool to renovate a crumbling mansion. The protagonist was a grandmother. The story, such as it was, involved a family mystery buried somewhere in the estate. The name, Merge Mansion, sounded like a retirement-home hobby.
Secondary transactions and shareholder swaps increasingly reflect broader changes as startup ecosystems adapt to longer timelines and delayed IPO markets.