AI-powered attacks and shadow AI adoption are creating new security risks inside the browser. Push Security explains why browser visibility is becoming critical for both threat detection and AI governance.
Get the latest international news and world events from around the world.
Zoltan Istvan: The Transhumanist Wager Is A Choice We’ll All Have To Make
Thirteen years ago, I sat down with a writer who had just published his first novel.
It was Zoltan Istvan’s very first media interview as a book author.
The book was The Transhumanist Wager. The question behind it was simple and almost unbearable: what would you do, and what would you give up, to live forever?
I loved half of it. I argued with the other half. That tension is exactly why I think it still matters.
Zoltan built his story out of Plato and Nietzsche, out of Thomas More’s Utopia and Zen Buddhism, then wrapped it all in an Atlas Shrugged plot of lone heroes and evil states. The philosophy is sophisticated. The framing is stark. The contradictions are not a flaw. They are the point.
One line from our conversation has stayed with me for more than a decade:
Scientists found a giant magnetic “twist” hidden inside the Milky Way
A hidden magnetic twist inside the Milky Way may rewrite what scientists know about how our galaxy is held together. Astronomers have uncovered a strange magnetic “flip” hidden inside the Milky Way. Using a new radio telescope, researchers mapped the galaxy’s magnetic field in unprecedented detail and discovered that a mysterious reversal in the Sagittarius Arm cuts diagonally across space. The finding could reshape how scientists understand the structure and future evolution of our galaxy.
For hundreds of years, astronomers have studied the night sky in an effort to understand the forces shaping the universe. One of the most important, yet invisible, forces inside the Milky Way is its magnetic field. Now, researchers at the University of Calgary are producing one of the clearest views yet of that hidden structure.
“Without a magnetic field, the galaxy would collapse in on itself due to gravity,” says Brown, a professor in the Department of Physics and Astronomy at the University of Calgary.
AI system automates scientific software design, outperforming human-written code in key benchmarks
A research team at Google co-led by Michael Brenner, Catalyst Professor of Applied Mathematics and Physics at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and Google research scientist, has produced a new artificial intelligence system that can automatically write scientific software programs that surpass the performance of human-written programs. The paper is published in the journal Nature.
How the ERA system came together The system is called Empirical Research Assistance (ERA), and the project was co-led by Brenner and Shibl Mourad from Google DeepMind. Harvard Ph.D. students Qian-Ze Zhu, Ryan Krueger, and Sarah Martinson contributed as Google student researchers while working in Brenner’s group. The research was done in Brenner’s capacity as a Catalyst Professor, a position established by the University to enhance relationships between academia and the private sector by supporting senior faculty in research roles at external companies.
Across modern science, customized software is constantly used to test specific hypotheses or interpret complex data. The authors refer to this type of computer program as “empirical software”—a program whose sole purpose is to maximize how well it does on a scientific task, like making weather predictions or forecasting hospitalizations during a disease outbreak. Any problem that can be expressed as a numerical value—its “score”—is called a scorable task.
Building a hill-climbing machine: Launching seven new MAI models
The title’s “hill-climbing machine” refers to Microsoft’s iterative, scientifically rigorous engineering framework. By tightly linking clean data pipelines, specialized training infrastructure, and reinforcement learning environments, they have created an optimization loop designed to steadily “climb” toward higher capabilities as compute scales.
Today we are announcing a family of seven new models developed in-house at Microsoft AI. Beyond these models, we’re building a superintelligence lab – a system and an approach we believe will define the next phase of AI.
This is an extraordinary time in technology. The compute used to train frontier models has increased by a factor of one trillion. Now we expect another thousand-fold increase over the next three years, which in turn means more advanced capabilities, and the continued rollout of ever more effective AI.
This epic compute ramp will change the nature of work, business and daily life. We all have to prepare for this reality. Our job at MAI is to help you do this – to push the frontier, and to build a hill-climbing machine to keep you at the frontier.
Explosive cytotoxicity of ruptoblasts bridges hormone surveillance and immune defense
Ruptoblasts are a cytotoxic glandular cell type that undergoes explosive cell death, ruptosis, releasing potent, broad-spectrum agents capable of killing nearby cells. Ruptosis is triggered by the hormone and cytokine activin and driven by ER-derived calcium release amplified by the actomyosin cytoskeleton.
LLMs help robots understand vague instructions and focus on key details
Imagine working at a warehouse or office sometime in the near future, and you’re asked to help a new trainee learn the basics of their job. The catch: It’s a robot. To teach them, you might want to play a game of “show and tell”—that is, physically showing how to do something a few different ways, while also explaining what you’re doing.
Let’s say you asked the robot to place some coffee on your desk without disturbing you during a Zoom call. You’ll prefer that the robot doesn’t get close to you and the laptop so that it doesn’t interrupt your meeting. To enable this behavior, the robot should be trained with data that clearly demonstrates the full task. Computer scientists have attempted to explain manipulation tasks to robots by recording lots of physical demonstrations or writing extensive directions. But if you don’t have both, the machine is likely to misunderstand what it needs to do.
It’s laborious for humans to do all that showing and telling, so researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have automated the process of teaching a robot, while clarifying instructions automatically and using nearly five times less demonstration data.
Rare meteorite provides evidence of giant early planet
Four-and-a-half billion years ago, a massive world—possibly as big as the moon or even Mars—orbited our sun before crashing into another celestial body and shattering into rubble. Now, in a paper published in the journal Earth and Planetary Science Letters, scientists report the first definitive evidence that this lost planetary embryo (protoplanet) existed. Its unique geological makeup challenges long-held assumptions about how planets evolve.
“It’s incredible to think there was once a world this large,” said Aaron Bell, an assistant research professor in the Department of Earth Science at the University of Colorado Boulder. “We only know it existed because a few fragments of it happened to land on Earth. These meteorites preserved evidence of a completely different pathway through which early planets developed.”
What gave away the lost world’s secret was a piece of its debris uncovered on Earth in the Sahara Desert, known as the Northwest Africa (NWA) 12,774 angrite meteorite.