AI Finds 160,000 New Viruses & Microsoft’s Nuclear-Powered AI

A LARRUZ YouTube video summarizing recent AI-driven developments: an international research team used an AI tool called lucaprot to identify 161,979 new RNA viruses in a Cell paper, revealing vast viral diversity; major tech firms (Microsoft, Amazon, Google, Meta) are securing nuclear power to meet growing AI/data-center energy needs (IDC projects data-center electricity will reach ~857 TWh by 2028); and leading AI researchers tied to Big Tech received Nobel Prizes while OpenAI expands in Europe and pursues large fundraising. The piece frames these as transformative but also raising ethical, power and safety concerns.
Questions this video answers
161,979
Cell
lucaprot
University of Sydney; Alibaba Cloud / Apsara Lab; Sun Yat-sen University (China).
Data-center electricity use expected to double between 2023 and 2028, reaching 857 terawatt-hours (TWh).
Chapters
All right, so you won't believe what's happening right now in the world of AI and science. I'm talking mind blowing discoveries. Artificial intelligence doing what we thought only humans could, and the unexpected return of something as controversial as nuclear power. Stick around, because today we're diving deep into AI that's discovering hundreds of thousands of new virus species. Microsoft betting on nuclear power to fuel its AI revolution and, and even Google researchers snagging Nobel prizes. So let's get into it, okay? Let's kick things off with a discovery that's got the entire scientific community buzzing. Researchers from Australia and China just used AI to identify over 160,000 new virus species. Yes. 161,979 new RNA viruses have been discovered using a machine learning tool. And this could just be the beginning. Now, what's crazy is that this is the largest virus discovery ever made in a single study. The research team, led by scientists from the University of Sydney, Alibaba Clouds, Apsara Lab and Sun Yat Sen University in China, dropped this bombshell in a study published in Cell, a big name journal in the science world. So how did they pull this off? Enter lucaprot, an AI powered tool they created. It basically sifts through huge amounts of genomic data, which is like an encyclopedia of genetic sequences and organizes all that information to identify viruses. A lot of these viruses were already sequenced. They were out there in the databases, but no one had connected the dots. Think of it like a bunch of puzzle pieces lying around. And Leuka Prot is the super smart AI that finally puts them together. And many of these viruses were hiding in some of the planet's most extreme environments. Places like hot springs, hydrothermal vents, and even the atmosphere. So. So, yeah, viruses are pretty much everywhere. But this doesn't just mean more viruses to worry about. They also play key roles in our ecosystems. It's all part of the bigger picture of how life works on Earth. Now, this is just the start. Edward Holmes, one of the lead researchers, said this study scratches the surface. There are millions more species out there waiting to be discovered. With AI tools like Leucaprot, they believe we can also uncover new bacteria and parasites. This changes the game in a huge way for biology, health, and even how we understand life itself. And fun fact, the AI tool lucaprot didn't just magically invent new data. It took what was already there and organized it in a way that humans simply couldn't do before. So it's not about replacing scientists, but rather supercharging Their ability to understand the natural world. Holmes and his team aren't done yet. They're already working on training Luca Prat to find even more viral diversity. The potential applications of this are massive, from discovering new pathogens to possibly unlocking new treatments, or even predicting viral outbreaks. In short, AI is making our world both more fascinating and a little scarier. But hold up. Before you think AI is only doing cool stuff like discovering viruses, there's a catch. AI, for all its awesomeness, is a massive energy hog. And I mean massive. Data centers where all these AI computations are happening are, are consuming insane amounts of electricity. According to International Data Corporation, data center electricity use is expected to double between 2023 and 2028, reaching a staggering 857 terawatt hours. That's more energy than some countries use now. That's where nuclear power comes back into the picture. And surprisingly, it's not the villain you might remember. Microsoft, of all companies, is breathing new life into one of the most infamous nuclear plants in history. Three Mile island, the site of the worst nuclear accident in the US back in 1979. But times have changed and nuclear energy is now seen as a cleaner alternative, since it doesn't emit greenhouse gases. And Microsoft, along with other tech giants like Google, Amazon and Meta, is all in on nuclear power. To fuel their AI driven futures. Microsoft has struck a deal to restart one of the dormant reactors at Three Mile island, the one that wasn't affected by the meltdown. And they're signing on to buy its electricity for the next 20 years. That's a massive bet on nuclear energy's role in the AI revolution. And it doesn't stop there. Microsoft has also partnered with Ontario Power Generation in Canada and Helion, a nuclear fusion startup. Amazon's in on the action too. Earlier this year, they dropped a cool $650 million for data centers powered by Pennsylvania's Susquehanna nuclear plant. And get this, they're not just going for the traditional nuclear reactors. Companies like Equinix and Oracle are looking into small modular reactors. New technology that could provide more efficient, decentralized power for data centers without needing to refuel for a decade. Imagine AI powered data centers running on cutting edge nuclear tech that never has to stop. And that's the future. With AI demand skyrocketing, tech giants are securing long term power supplies. And nuclear energy is making a serious comeback. Now. Speaking of AI making history, Google researchers have just won the Nobel Prize. So Nobel Prizes in chemistry and physics were awarded to AI pioneers this week. First up, Demis hassabis and John jumper from Google. DeepMind, along with US biochemist David Baker, grabbed the chemistry Nobel for decoding microscopic protein structures using AI. This is huge because proteins are the building blocks of life, and understanding their structures could unlock everything from new medicines to better understanding diseases. Then there's Geoffrey Hinton, one of the godfathers of AI, who won the Nobel in physics for his groundbreaking work in machine learning. Hinton's work laid the foundation for the AI explosion we're seeing today. And although he quit Google last year to speak more freely about the dangers of AI, his impact on the field is undeniable. But not everyone is happy about these awards. Some critics argue that what Hinton and the others did wasn't really physics or chemistry in the traditional sense. The Nobel committee had to get creative in how they awarded these prizes, since there isn't a Nobel for computer science or AI. Regardless of the controversy, this speaks to a larger issue. Big Tech is now leading the charge in scientific breakthroughs, outpacing traditional academia. Google, Microsoft, and Amazon have the resources and talent to push AI research to new heights, but it's causing a lot of debate about how much power these companies should have over the future of science. Hinton himself has expressed regret about where AI is headed, particularly the existential threat of AI becoming smarter than humans. He's not alone in his concerns, but the fact remains Big Tech is reshaping science and pushing the boundaries of what we thought was possible. Now let's shift our focus to OpenAI, the company behind ChatGPT, and how they're expanding their reach in Europe. OpenAI confirmed that it's opening new offices in Paris and Brussels this year as part of its global expansion plans. This is in addition to new offices in New York and Singapore. So why Europe? Well, for one, it's a strategic move to collaborate more with European developers, companies and organizations. Europe has been at the forefront of pushing for AI regulation and ethical development, so having a presence there helps OpenAI navigate those waters while also tapping into Europe's tech talent. Plus, OpenAI is currently raising a whopping $6.5 billion in a funding round to that would value the company at over $150 billion. That's some serious cash, and it shows just how big the stakes are in the AI race. So to wrap it all up, AI is literally changing everything from discovering tens of thousands of viruses and revolutionizing scientific research to powering data centers with nuclear energy and winning Nobel prizes. AI is taking over, but it's also raising big questions about power, ethics, and where we're headed as a society. What do you guys think? Is this AI takeover exciting or are we playing with fire? Let me know in the comments below. Don't forget to hit that like button and subscribe for more updates on how AI is reshaping our world.










