Watch: South Bay mountain lion and cubs caught on camera
Vishal Subramanyan and Cynthia Cross had spent three years working to film and photograph mountain lions in the Diablo Range, according to Subramanyan. In February, one of their cameras set up with a motion sensor in the East Foothills of San Jose captured not only a mountain lion, but a mom with her three young cubs.
Study reveals ranging diet of San Francisco coyotes
The precious video shows the mom nursing her cubs as well as picking one up while the two others prance around the wooded area. The mountain lion family appeared in front of the camera for 20 minutes in total, Subramanyan said, calling it 'one of the most special moments we've ever captured.'
'Mountain lions are obviously an incredible species,' Subramanyan told KRON4. 'As an apex predator in the Bay Area, they play a really important role in top-down ecological effects. So seeing a healthy, thriving breeding population of mountain lions gives me a lot of hope, especially with a lot of negative news we see about our environment today.'
Subramanyan is a UC Berkeley graduate with degrees in ecology and statistics, and has worked with organizations such as the California Academy of Sciences, Peninsula Open Space Trust, and Midpeninsula Regional Open Space District.
Approximately 4,500 mountain lions are roaming across the state, according to the UC Davis California Mountain Lion Project.
Copyright 2025 Nexstar Media, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Tom's Guide
31-07-2025
- Tom's Guide
AI models can secretly influence each other — new study reveals hidden behavior transfer
A new study from Anthropic, UC Berkeley, and others reveals that AI models may also be learning from each other, via a phenomenon called subliminal learning, not just from humans. Not exactly gibberlink, as I've reported before, this communication process allows one AI ('teacher') to pass behavioral traits, such as a preference for owls, or even harmful ideologies, to another AI ('student'). All of this influencing is done through seemingly unrelated data, such as random number sequences or code snippets. In experiments, a teacher model was first tuned with a trait (e.g., loving owls) and then asked to generate 'clean' training data, such as lists of numbers, with no mention or reference to owls. A student model trained only on those numbers later exhibited a strong preference for owls, compared to control groups. The effect held even after aggressive filtering. The same technique transmitted misaligned or antisocial behavior when the teacher model was deliberately misaligned, even though the student model's training data contained no explicit harmful content. The study seems to indicate that filtering isn't enough. Most AI safety protocols focus on filtering out harmful or biased content before training. But this study shows that even when the visible data looks clean, subtle statistical patterns, completely invisible to humans, can carry over unwanted traits like bias or misalignment. Get instant access to breaking news, the hottest reviews, great deals and helpful tips. And, it creates a chain reaction. Developers often train new models using outputs from existing ones, especially during fine-tuning or model distillation. This means hidden behaviors can quietly transfer from one model to another without anyone realizing. The findings reveal a significant limitation in current AI evaluation practices: a model may appear well-behaved on the surface, yet still harbor latent traits that could emerge later, particularly when models are reused, repurposed, or combined across generations. For AI developers and users alike, this research is a wake-up call; even when model-generated data appears harmless, it may carry hidden traits that influence future models in unpredictable ways. Platforms that rely on outputs from other models, whether through chain-of-thought reasoning or synthetic data generation, may unknowingly pass along biases or behaviors from one system to the next. To prevent this kind of 'behavioral contamination,' AI companies may need to implement stricter tracking of data origins (provenance) and adopt safety measures that go beyond simple content filtering. As models increasingly learn from each other, ensuring the integrity of training data is absolutely essential. Follow Tom's Guide on Google News to get our up-to-date news, how-tos, and reviews in your feeds. Make sure to click the Follow button.
Yahoo
22-07-2025
- Yahoo
Researchers pack a "quantum light factory" into a 1mm² CMOS chip — combines photonics, electronics, and quantum hardware with traditional silicon manufacturing that can achieve mass scale
When you buy through links on our articles, Future and its syndication partners may earn a commission. Researchers from Boston University, UC Berkeley, and Northwestern University have created something that sounds straight out of a sci-fi movie: a 'quantum light factory' squeezed onto a 1 mm² silicon chip. Built using a standard 45 nm CMOS manufacturing process—the same kind used for standard x86 and ARM processors—this breakthrough brings quantum hardware a step closer to the world of mass production. The work, published in Nature Electronic, could pave the way for scalable quantum computing that doesn't require exotic setups, instead relying on mass production techniques that we already employ today. Think of this chip as a prototype for a future quantum factory line. It packs 12 tiny silicon loops, called "microring resonators," each acting as a generator of photon pairs with special quantum properties. These photon pairs are the lifeblood of many quantum technologies, but producing them usually requires fragile lab setups. Here, they're generated directly on a chip no bigger than a fingernail. What makes this remarkable is that the chip doesn't just produce quantum light; it's more about keeping that light stable. Microring resonators are powerful but temperamental—small temperature changes or manufacturing quirks can throw them out of tune, halting the photon flow. To solve this, the researchers built a feedback system directly into the chip; each resonator has a tiny photodiode to monitor its performance, along with miniature heaters and control circuits that adjust it on the fly. This self-tuning approach means all 12 resonators can work together in perfect sync, without the bulky stabilization equipment usually needed. 'This is a small step, but an important one,' said Miloš Popović, associate professor at Boston University and one of the senior authors. 'It shows we can build repeatable, controllable quantum systems in commercial semiconductor foundries.' That's the real story here—this isn't just a niche, bleeding-edge lab demo, but proof that quantum chips can be made with the same industrial techniques used to build CPUs and GPUs. Of course, quantum computing is nowhere near the maturity of standard semiconductors powering our devices today, but this is a step closer to that eventuality. Let's talk about the most important revelation. The team's choice of CMOS (Complementary Metal-Oxide-Semiconductor) is a game-changer. CMOS is the backbone of modern electronics, used by companies like TSMC to mass-produce everything from smartphones to supercomputers. While the 45 nm node used here isn't cutting-edge, it's proven, cost-effective, and compatible with the vast infrastructure of silicon manufacturing. The chip was built using a platform co-developed with GlobalFoundries and Ayar Labs, a company already leading the charge in optical interconnects for AI and high-performance computing. This overlap with the AI world is no coincidence. Nvidia CEO Jensen Huang recently called out microring resonators—like those on this chip—as key components for scaling AI hardware via optical connections. This new research shows that the same photonics technology could also unlock scalable quantum systems. It's not hard to imagine a future where quantum and AI hardware share similar silicon platforms. Moreover, Nvidia is already heavily investing in this field, so we can only expect development to ramp up. The term 'quantum light factory' isn't just for flair, either. Just as classical chips rely on streams of electrons and optical networks depend on laser light, future quantum technologies will need a steady supply of quantum light. By proving that these quantum light sources can be built, stabilized, and replicated on silicon, the team has shown that quantum hardware can move beyond one-off experiments and into something that can be scaled like traditional computing. Some of the researchers involved are already taking this expertise into industry roles. Team members have joined companies like PsiQuantum, Ayar Labs, and Google X, all of which are betting heavily on photonic and quantum technologies. It's another sign that this field is moving rapidly from academic research to real-world products, even if they're sometimes more fun-oriented than revolutionary. Backed by the National Science Foundation's Future of Semiconductors (FuSe) program, the Packard Fellowship, and the Catalyst Foundation, the project shows how far interdisciplinary collaboration can go. Photonics, electronics, and quantum optics are worlds apart, but this chip proves they can be brought together on a commercial platform. If Intel's 4004 microprocessor marked the start of mass-produced computing power, this 1 mm² quantum light factory could be remembered as the first step toward mass-produced quantum hardware. What once needed an entire lab bench now fits onto a silicon wafer—and that's a leap worth paying attention to. Who knows? Maybe a decade from now, you'll be seeing a new TSMC competing for quantum computing excellence, especially now that there's already an operating system for it out there. Follow Tom's Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.
San Francisco Chronicle
22-07-2025
- San Francisco Chronicle
Satellites launching from California could help scientists understand disruptive space weather
Two twin satellites — part of NASA's TRACERS mission — set to launch tomorrow at Vandenberg Space Force Base will help scientists better understand space weather. Researchers expect the mission to provide new insights into how charged particles from the sun plow into Earth's magnetic shield. Such interactions underlie the dazzling northern lights, like the brilliant light shows that swept large swaths of California in 2024. 'What we will learn from TRACERS is critical for the understanding and eventually the predicting of how energy from our sun impacts the Earth,' said Joe Westlake, director of NASA's heliophysics division, during a briefing July 17. 'It's going to help us keep our way of life safe here on Earth.' The explosion of energy can also wreak havoc on satellites and power grids. Data collected by the TRACERS mission could aid future aurora forecasts and also inform preparations ahead of damaging geomagnetic storms. Two identical satellites will capture data on how a steady stream of charged particles from the sun, also known as solar wind, enters the near-Earth environment. This process is known as magnetic reconnection. High solar wind speeds can bring geomagnetic storms while slow wind speeds bring calm space weather. Fast solar winds can also cause auroral displays. The most expansive displays, like those above California in 2024, are typically associated with coronal mass ejections, large expulsions of charged particles and magnetic fields from the sun. Solar winds carry the sun's embedded magnetic field, reaching speeds over one million miles per hour. Earth's magnetosphere, a region dominated by the planet's magnetic field, acts as a protective bubble deflecting the solar wind and shielding Earth from solar radiation. The twin satellites, each slightly larger than a washing machine,will fly through a narrow region in Earth's magnetic field known as a polar cusp, where solar wind funnels into Earth's atmosphere. Over a year, the spacecraft will make 3,000 passes through the northern polar cusp. The reason for two satellites, rather than just one, is to probe how magnetic reconnection occurs in space and time. The two TRACERS spacecraft will follow each other in orbit, creating closely spaced snapshots that will help scientists tease apart how quickly changes in plasma, or charged particles from the sun, occur. The two satellites carry scientific equipment provided by UC Berkeley, UCLA and other institutions. UC Berkeley's instruments sense electricity within the magnetized plasma of near-Earth space, said David Miles, an associate professor of physics and astronomy at the University of Iowa and principal investigator of the TRACERS mission, by email. UCLA's instruments work like extremely sensitive compasses and measure the strength and direction of the local magnetic field. 'These instruments from Berkeley and UCLA are crucial for decoding the complex interactions in our planet's space environment and the magnetic reconnection process that couples in energy from the solar wind,' Miles said. Decoding such interactions is important for life closer to the surface. The historic storm in May 2024 that led to aurora viewings in California impacted air travel: Transoceanic flights were rerouted due to high-frequency radio loss, Westlake said. The event also caused roughly $500 million in economic losses in the agriculture sector, due to precision GPS issues, Westlake added. 'TRACERS joins the fleet of current heliophysics missions that are actively increasing our understanding of the sun and space weather and how to mitigate these impacts,' Westlake said.
