Large investors hold long-term positions and collect rent through hedging transactions, effectively lowering their average cost basis. For them, Bitcoin becomes an asset that generates interest; as long as their total position size doesn't decrease, over time the cost basis becomes so low that they almost break even. Many ordinary people mistakenly view large investors simply as short sellers; they are more like landlords collecting rent. This logic was consistent with some of the previous domestic industrial commodity spot exchanges. Large investors held physical assets long-term and collected rent through warehouse receipts, financing, and delivery, continuously lowering their average cost basis. The forward spread in CME Bitcoin futures essentially reflects the market's pricing of borrowing costs during that period. Collateral fees are essentially a reflection of funding rates. Whether the forward structure is at a discount or premium, it reveals the market's expectation of holding costs, which is completely consistent with the rent-collecting logic of large physical investors.

A large AI data center project in Utah’s Hansel Valley, called the Stratos Project, is getting significant attention. At full capacity, it would use up to 9 gigawatts of electricity, more than twice the total power consumed by Utah today. The facility will generate its own electricity via on-site natural gas plants, and all of it turns into heat. A Utah State University physics professor calculated that the daily heat output equals the energy from about 23 Hiroshima atomic bombs. But this heat is released slowly over 24 hours across a 62-square-mile area through large cooling systems, unlike a real bomb explosion. For comparison, the sun shining on the same desert land provides about twice as much energy daily. Local residents claim the project could raise nighttime temperatures by 8 to 12 degrees Fahrenheit, use large amounts of water in a dry area near the Great Salt Lake, and add significant pollution that heats the planet. County leaders recently approved the project, but opponents are pushing for a public vote that could block it but many people support it because it will create thousands of jobs

A large data center in Fayetteville, Georgia, used more than 29 million gallons of water over many months while the local water company was unaware and did not send a bill at first. Residents in the Annelise Park neighborhood complained about low water pressure, which revealed the issue. The QTS data center sits on 615 acres with two large water pipes that were improperly installed, as one was added without notifying the water company. Most water was used during construction for pouring concrete and controlling dust. After the error was discovered, QTS received a bill for about $150,000 The company says that once fully operational, it will use a closed-loop cooling system requiring very little water each month. As a result, the Fayetteville City Council voted to ban all new data centers in the city.

Five large book companies and writer Scott Turow have filed a lawsuit against Meta and Mark Zuckerberg. They say Meta took millions of protected books, textbooks, and articles from illegal websites including LibGen, Sci-Hub, and Anna’s Archive, then used this material to train its Llama AI models. The lawsuit claims Zuckerberg personally approved and encouraged the copying. It also says Meta removed or changed the copyright notices on the works. The companies argue this was planned large-scale theft, not just normal web scraping. They want the court to make Meta pay money and stop using their books. This is one of many lawsuits asking if AI companies can use protected material to train their systems without permission.