Critical Mass: Unearthing the secrets of dark energy
'The physical mechanisms responsible for the accelerating expansion of the universe is [sic] one of the most important unsolved problems in physics and, arguably, all of science,' according to a team of researchers called the Dark Energy Spectroscopic Instrument (DESI) Collaboration, who have just announced a striking new discovery.
Dark energy isn't what's currently emanating from the White House, but rather, the name astronomers and cosmologists give to whatever it is that is causing our expanding universe to accelerate: to expand ever faster. This cosmic acceleration was first seen in 1998 from observations of extremely distant supernovae, but there is no agreed explanation for it.
Specifically: the amount of dark energy doesn't stay constant over time. To give you some orientation, if anyone were to make a claim like that for the ordinary matter or energy in the universe, the impact on physics would be seismic. It's one of the most fundamental tenets of physics that the amount of regular matter/energy (Einstein's E=mc2 shows that the two are related) in the universe is constant. But for dark energy we have no expectations because we don't know what it is. The discovery that dark energy seems to be dynamic – changeable – should help to narrow down the possibilities.
This is no small matter for cosmologists. It is embarrassing enough that the amount of ordinary matter in the universe is apparently outweighed by about five times as much of the stuff called dark matter, which seems required to explain why galaxies don't fly apart and for other astrophysical reasons, but which remains totally unexplained because dark matter seems only to affect light and ordinary matter via its gravity. Efforts to detect dark-matter particles have produced nothing at all for decades. But then dark energy comes along, of which there is three times as much in the universe as the energy embodied by all the ordinary and dark matter put together. In other words, we don't actually know what about three-quarters of the universe is made of.
Dark energy manifests itself as a kind of repulsive force that opposes gravity. The expansion of the entire cosmos is just what Einstein's theory of gravity – general relativity – predicts, and it means that all galaxies are receding from each other in all directions. (The usual analogy asks us to imagine them drawn on the surface of a balloon that is being blown up.) The gravitational attraction between galaxies pulls the other way, but not enough to curb the expansion. But if, as the 1998 discovery showed, the expansion is getting steadily faster, this implies the existence of some repulsive force that is speeding up the recession.
One way to think about dark energy is in terms of the so-called cosmological constant. This was an idea Einstein introduced when he realised that his equations of general relativity predicted the universe was expanding – because at that point we didn't know that it was, and so Einstein figured his theory needed some added factor to prevent it. Once Edwin Hubble revealed the cosmic expansion in 1929, the cosmological constant seemed otiose. But adding it back into the equations can capture the accelerating influence of dark energy.
But that doesn't help to explain where it comes from. One option is supplied by quantum physics, which predicts that empty space should be bubbling with 'virtual' particles that pop briefly in and out of existence, creating a kind of pervasive background of energy. The trouble is that this should produce way more energy than we need to explain dark energy, creating the problem of having to explain how most of that extra energy gets cancelled out but a tiny bit remains.
There's no room to talk about the other ideas proposed to explain dark energy. But all are speculation without data. Now the DESI team has collected three years' worth of observations on distant galaxies and other astrophysical objects – the largest ever 3D map of the universe – to follow the cosmic expansion over time. (Because of the finite speed of light, the further we look out into the universe, the further back in time we're seeing.) Their first results were announced last year, but now they can conclude with more confidence that, when the findings are combined with other observations, they imply that dark energy has been growing weaker over time. If so, it's less likely that ultimately dark energy will rip all of spacetime apart.
There's no good theoretical explanation for dynamical dark energy, and it conflicts with what is considered the best cosmological model so far, called the lambda cold-dark-matter model (which assumes a cosmological constant). For theoreticians, that's actually great news: there's concrete data to sink their teeth into.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
The Guardian
4 days ago
- The Guardian
How the earth shook for nine days and nobody knew why
An unprecedented planetary-scale seismic event caused the earth to vibrate for nine days straight back in 2023, but the reason why was unclear. Scientists initially had more questions than answers, labelling the event an unidentified seismic object and undertook a mammoth scientific collaboration across multiple countries and institutions to get to the bottom of what really happened. Josh Toussaint-Strauss looks into the mystery at the heart of this scientific investigation
Daily Mail
12-06-2025
- Daily Mail
You're cooking your steak wrong! Scientists reveal the Einstein-inspired formula for a perfect fillet
With Father's Day approaching, some Brits may be planning on rustling up a nice dinner to treat their Dad. Now, scientists have revealed the formula for cooking the perfect steak – and it's all down to how heat moves through meat. The equation, inspired by Albert Einstein, features one very important factor – the thickness of the meat. And since the total cooking time can vary from three to 14 minutes, it's one equation that could turn out very useful in the kitchen. It has been devised by Professor Rob Appleby, a physicist from the University of Manchester and a trained chef. He has utilised a formula first worked out by Einstein - L² = 4Dt - originally used to describe the motion of particles. In the formula L represents the thickness of the steak, T is the time for the steak to cook and D is the diffusion coefficient, which is the speed at which heat moves through the fillet. Professor Appleby has used the equation to create a simplified timing chart to help home chefs master steak – so, will you put it to the test? 'Steak science is surprisingly precise,' Professor Appleby said. 'It's not the weight of the steak that matters, but its thickness. 'Heat takes longer to travel through thicker meat, and Einstein's formula helps us understand exactly how long it takes.' He explained that the time it takes to cook a steak increases with the square of its thickness. That means doubling the thickness of a steak doesn't just double the cooking time - it can quadruple it. This insight busts one of the biggest BBQ myths – that a steak twice as big takes twice as long to cook. By using the formula and a simple temperature probe, home cooks can avoid overcooking, undercooking or the dreaded burst-outside, raw-inside steak disasters. 'A 2cm-thick steak might take just four minutes in total, while a 4cm steak could take over 12 minutes,' Professor Appleby said. The timing is perfect for Father's Day weekend, when some people might be firing up their BBQs. 'People are always nervous about steak, especially on a BBQ, but this takes the guesswork out,' Josh Novell, from Polhill BBQs, said. 'You don't need to be a physicist; just have the right tools and the right method. 'When you're cooking over flames or coals, heat zones vary and it's easy to burn the outside before the centre's ready. 'But by understanding how heat travels through the steak, you avoid solely relying on instinct.' Steak cuts explained RUMP: The rump steak is a cut of beef taken from the hindquarter of the animal, specifically the upper part of the leg. This area is a hard-working muscle, which means that the meat is leaner than other cuts, but also incredibly flavourful. RIBEYE: Ribeye is a classic steak cut, known for its exceptional flavour and marbling. Cut from the rib section of the cow, it is one of the most tender and juicy cuts available. The ribeye is characterized by its distinct eye of marbling, which ensures a rich and juicy flavour when cooked. FILET MIGNON: Filet Mignon is known as the king of steaks, thanks to its extreme tenderness and delicate, buttery flavour. This steak comes from the tenderloin, which is located just below the spine of the cow. Because this muscle isn't used as much during the cow's lifetime, it is one of the most tender cuts available. FILLET: The fillet steak, or tenderloin steak, is a cut of beef that comes from the tenderloin – a long, slender muscle located beneath the ribs and along the spine of the cow. This muscle does very little work, resulting in a cut that is incredibly tender and delicate. Due to its premium quality and limited availability (only about 4-6 pounds of tenderloin per cow), fillet steaks are often considered a luxury item and can command a higher price at your butcher than other cuts. SIRLOIN: Sirloin is a versatile steak cut, characterized by its lean, but flavourful meat. It comes from the rear of the cow, just before the rump. Because this muscle is used frequently during the cow's lifetime, sirloin can be less tender than other cuts but still has a unique and robust taste. T-BONE: The T-bone steak is a combination of two steaks: the tenderloin and the strip steak. It is characterized by a T-shaped bone in the centre, which separates the two cuts of meat. T-bone is known for its rich beefy flavour and tenderloin's buttery, tender texture. PORTERHOUSE: Porterhouse is similar to T-bone in that it consists of two steaks separated by a Tshaped bone. However, it has a larger portion of the tenderloin, making it the ideal steak cut for those who want the best of both worlds. It's a big steak, perfect for sharing or perfect for a large appetite. NEW YORK STRIP: The New York Strip is a popular steak cut known for its juicy and flavourful meat. This steak comes from the beef loin and is characterized by its leaner, yet still tender, texture. Because of its consistency, it can take on a range of marinades and seasonings, making it a versatile steak. FLANK: Flank steak is a lean cut of meat that comes from the abdominal muscle of the cow. While not as tender as other cuts, it is known for its rich flavour and versatility in cooking. Flank steak is ideal for grilling or pan-frying, and it makes an excellent addition to salads or tacos.
Geeky Gadgets
30-05-2025
- Geeky Gadgets
Andrew Ng on the Future of AI : From Assistants to Autonomous Collaborators
What does the future of artificial intelligence look like when the boundaries between human and machine collaboration blur, and AI agents evolve to make decisions autonomously? At the LangChain Interrupt 2025 conference, Andrew Ng, one of the most influential voices in AI, tackled this question head-on. In a candid and thought-provoking exchange, Ng explored the fantastic potential of AI agents, from their ability to automate complex workflows to their role in reshaping industries like customer service, software development, and entrepreneurship. With frameworks like LangChain and the Model-Context-Protocol (MCP) leading the charge, Ng painted a vivid picture of a world where AI systems don't just assist but actively collaborate with humans to solve high-stakes problems. Yet, as he pointed out, this vision isn't without its challenges—designing scalable, efficient systems while maintaining ethical and technical integrity remains a daunting task. During this interview, viewers will gain a front-row seat to Ng's insights on the spectrum of autonomy in agentic systems, the critical skills developers need to thrive, and the emerging trends poised to redefine the AI landscape. From the rise of multi-agent collaboration to the untapped potential of voice applications, Ng's commentary offers a roadmap for navigating the complexities of an AI-driven future. Whether you're a developer, entrepreneur, or simply curious about the next wave of innovation, this dialogue with Andrew Ng—curated by LangChain—promises to challenge assumptions and spark new ideas. As Ng himself suggests, understanding these shifts isn't just about keeping up; it's about staying ahead in a world where AI is no longer a tool but a partner. Andrew Ng on AI Evolution Understanding Agentic Systems: A Spectrum of Autonomy Ng underscored a pivotal shift in the AI domain: the focus is no longer solely on defining AI agents but on understanding the varying degrees of autonomy within agentic systems. He explained that these systems operate along a spectrum, ranging from simple task automation to highly autonomous decision-making processes. Ng emphasized the importance of designing streamlined, linear workflows that simplify complex operations while maintaining efficiency. However, he acknowledged the inherent difficulty in breaking down intricate business processes into manageable, agent-driven tasks. These systems, he noted, are essential for automating repetitive tasks, improving productivity, and allowing businesses to scale operations effectively. Key Skills for AI Developers Ng highlighted the critical skills developers must cultivate to excel in the AI field. Proficiency in tools such as LangChain, LangGraph, and evaluation frameworks is essential for designing, debugging, and optimizing workflows. Beyond technical expertise, Ng stressed the importance of coding knowledge, not just for developers but also for non-technical professionals seeking to collaborate effectively in AI projects. He encouraged developers to hone their instincts for workflow optimization, a skill he deemed vital for creating efficient and scalable AI systems. This combination of technical and strategic skills, Ng argued, will empower professionals to thrive in the evolving AI landscape. Andrew Ng Explains the Future of AI Collaboration at LangChain 2025 Watch this video on YouTube. Stay informed about the latest in Agentic systems by exploring our other resources and articles. Voice Applications: Enhancing User Experience Voice-based interactions are emerging as a powerful tool for reducing user friction and improving accessibility. Ng discussed the growing demand for low-latency responses in voice applications, particularly in enterprise and customer service settings. He highlighted innovative techniques such as pre-responses, which anticipate user needs to deliver faster and more seamless interactions. These advancements, Ng explained, have the potential to significantly enhance operational efficiency and user satisfaction. However, he acknowledged that achieving seamless integration and responsiveness remains a technical challenge that developers must address to unlock the full potential of voice technologies. MCP Framework: Simplifying AI Integration Ng introduced the Model-Context-Protocol (MCP) framework as a promising approach to standardizing data and tool integrations in AI systems. By reducing development complexity, MCP enables developers to streamline workflows involving multiple models and data sources. Ng emphasized that this framework can help build robust and efficient AI systems, but he also acknowledged persistent challenges, such as authentication and implementation barriers. These hurdles, he noted, require further refinement to fully realize MCP's potential. Despite these challenges, Ng highlighted MCP's ability to simplify the integration process, making it a valuable tool for developers working on complex AI projects. Multi-Agent Systems: Unlocking Collaboration Ng explored the potential of multi-agent systems to assist collaboration between AI agents. He explained that while current implementations are largely confined to single-team environments, where agents operate within controlled settings, the next frontier lies in allowing cross-team collaborations. These more complex dynamics, Ng argued, could unlock new possibilities for automating and optimizing workflows across diverse industries. He emphasized the importance of advancing multi-agent systems to achieve greater levels of coordination and efficiency, which could transform how businesses approach large-scale operations. AI-Assisted Coding: Boosting Developer Productivity AI-assisted coding tools are reshaping the way developers approach their work. Ng highlighted the ability of these tools to automate repetitive coding tasks, allowing developers to focus on higher-level problem-solving and innovation. However, he cautioned that a strong foundation in programming fundamentals is essential to maximize the benefits of these tools. Ng encouraged broader adoption of AI-assisted coding across industries, citing its potential to accelerate innovation and reduce development timelines. By using these tools effectively, developers can enhance productivity and contribute to faster technological advancements. Entrepreneurship in AI: Strategies for Success Ng offered valuable advice for aspiring AI entrepreneurs, emphasizing two key factors for success: speed and technical expertise. He urged founders to prioritize rapid execution while developing a deep understanding of the technologies they are working with. According to Ng, innovation stems from a combination of technical depth and the ability to adapt quickly to changing market demands. He encouraged entrepreneurs to stay informed about emerging tools and methodologies, as this adaptability is crucial for maintaining a competitive edge in the fast-paced AI industry. Emerging Trends in AI Ng shared his perspective on the trends shaping the future of AI. He highlighted advancements in longer context memory for large language models (LLMs), which will enable AI systems to process and retain more information, improving their performance in complex tasks. Additionally, he pointed to underexplored areas such as voice applications and systematic evaluation frameworks as opportunities for innovation. Ng emphasized the importance of continuous learning and adaptation, urging developers and businesses to stay ahead by embracing new tools and methodologies. These trends, he noted, will play a crucial role in shaping the next generation of AI technologies. Media Credit: LangChain Filed Under: AI, Technology News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
