Latest news with #EPFL
Yahoo
6 days ago
- Health
- Yahoo
New study makes concerning discovery about farms where food is grown: 'A critical step in understanding the potential risks'
A recent study made a disturbing discovery about the ongoing problem of tire chemicals seeping into agricultural products. A recent study published in the Journal of Hazardous Materials found that additives used in tire manufacturing were present in common fruits and vegetables eaten in Switzerland. While further research is needed to understand the implications for human health, this discovery was described as "a critical step in understanding the potential risks associated with these contaminants." As explained in a report from the École Polytechnique Fédérale de Lausanne published by tires often "release additives such as antioxidants and vulcanizing agents" that can make their way to agriculture after dispersing through the air and being transported by runoff water. In addition to inhaling these chemicals, humans are exposed to them by ingesting contaminated food, researchers from the EPFL and the Swiss Federal Food Safety and Veterinary Office determined. "Our exposure to these additives is similar to that for other micropollutants," Florian Breider, who heads the Central Environmental Laboratory research group at the EPFL, said, per the university's report. "They're around us constantly, in every part of our environment. What we don't know is whether we need to introduce tighter controls, such as by phasing them out in tire manufacturing in favor of less toxic alternatives." As part of the study, scientists tested fruits and vegetables for 11 compounds typically found in tire additives and found traces of those compounds in 31% of the samples. The origin of the fruits and vegetables and whether they were organic made no difference in the findings, showing how wide the range of exposure is for the general public. "Since people's diets are broadly the same throughout Western Europe, there's a good chance that these findings are representative of consumers' exposure in neighboring countries," Breider said in the EPFL's report. This marks another growing concern for crops around the world. Many people are aware of the issue of fruits and vegetables being covered in pesticides. While useful for protecting crops, pesticides can have detrimental effects and can lead to an increased risk of significant health issues such as cancer and diabetes. The health effects of tire additives are still being explored, but any exposure to chemicals is unsettling. "We don't know how these compounds are metabolized by the human body," Breider said in the EPFL's report. "It's something we need to explore because we're all exposed to these compounds—and especially people in certain population groups, such as road workers." According to the EPFL's report, scientists in China are also looking into the subject and testing for the presence of tire additives in human blood and urine samples. Scientists in Switzerland are taking the broader approach of "exploring ways in which roads can be decontaminated to prevent tire additives from entering the environment." Do you worry about pesticides in your food? All the time Sometimes Not really I only eat organic Click your choice to see results and speak your mind. You can limit exposure to chemicals in various ways. Cooking at home as much as possible, using less plastic packaging, and growing your own produce are simple actions you can take to protect yourself. By increasing consumer awareness of these dangers, more people can put pressure on companies to phase out dangerous materials. Join our free newsletter for easy tips to save more and waste less, and don't miss this cool list of easy ways to help yourself while helping the planet.
Gizmodo
19-05-2025
- Science
- Gizmodo
AI Gets a Lot Better at Debating When It Knows Who You Are, Study Finds
A new study shows that GPT-4 reliably wins debates against its human counterparts in one-on-one conversations—and the technology gets even more persuasive when it knows your age, job, and political leanings. Researchers at EPFL in Switzerland, Princeton University, and the Fondazione Bruno Kessler in Italy paired 900 study participants with either a human debate partner or OpenAI's GPT-4, a large language model (LLM) that, by design, produces mostly text responses to human prompts. In some cases, the participants (both machine and human) had access to their counterparts' basic demographic info, including gender, age, education, employment, ethnicity, and political affiliation. The team's research—published today in Nature Human Behaviour—found that the AI was 64.4% more persuasive than human opponents when given that personal information; without the personal data, the AI's performance was indistinguishable from the human debaters. 'In recent decades, the diffusion of social media and other online platforms has expanded the potential of mass persuasion by enabling personalization or 'microtargeting'—the tailoring of messages to an individual or a group to enhance their persuasiveness,' the team wrote. When GPT-4 was allowed to personalize its arguments, it became significantly more persuasive than any human—boosting the odds of changing someone's mind by 81.2% compared to human-human debates. Importantly, human debaters did not become so persuasive when given access to that personal information. 'In the context of persuasion, experts have widely expressed concerns about the risk of LLMs being used to manipulate online conversations and pollute the information ecosystem by spreading misinformation, exacerbating political polarization, reinforcing echo chambers and persuading individuals to adopt new beliefs,' the researchers added. GPT-4 can argue with you, and given a set of facts about you, it may excel at convincing you to change your point of view, the researchers found. The team notes in the paper's discussion that LLMs have previously been criticized for generating and diffusing hate speech, misinformation, and propaganda; at scale, LLMs with users' personal information could be harnessed for malicious purposes. The team's research pairs nicely with a recent ChatGPT update that allows the model to remember more of users' conversations (with their permission), meaning that the AI can have access to a catalogue of information about its users. But there's also good news—or bad news—depending on how you see it. GPT-4 was very effective at persuading its opponents on less controversial issues, but with more entrenched positions (referred to in the research as 'opinion strength'), the bot had a harder time convincing humans to change their minds. In other words, there's no indication that GPT-4 would be any more successful than you are at the Thanksgiving debate table. What's more, the researchers found that GPT-4 tends to use more logical and analytical language, while human debaters relied more on personal pronouns and emotional appeals. Surprisingly, personalization didn't dramatically change GPT-4's tone or style—it just made its arguments more targeted. In three out of four cases, human participants could correctly identify their opponent as AI, which the researchers attribute to GPT-4's distinct writing style. But participants had a difficult time identifying human opponents as human. Regardless, people were more likely to change their mind when they thought they were arguing with an AI than when they believed their opponent was human. The team behind the study says this experiment should serve as a 'proof of concept' for what could happen on platforms like Reddit, Facebook, or X, where debates and controversial topics are routine—and bots are a very established presence. The recent paper shows that it doesn't take Cambridge Analytica-level profiling for an AI to change human minds, which the machines managed with just six types of personal information. As people increasingly rely on LLMs for help with rote tasks, homework, documentation, and even therapy, it's critical that human users remain circumspect about the information they're fed. It remains ironic that social media—once advertised as the connective tissue of the digital age—fuels loneliness and isolation, as two studies on chatbots found in March. So even if you find yourself in a debate with an LLM, ask yourself: What exactly is the point of discussing such a complicated human issue with a machine? And what do we lose when we hand over the art of persuasion to algorithms? Debating isn't just about winning an argument—it's a quintessentially human thing to do. There's a reason we seek out real conversations, especially one-on-one: To build personal connections and find common ground, something that machines, with all their powerful learning tools, are not capable of.
NDTV
29-04-2025
- Science
- NDTV
This Innovative "Robotic Cake" Was Lit By Edible Chocolate Batteries
Hyper-realistic cakes, burn-away cakes and similar cake innovations have been taking the world by storm for some time now. But have you ever come across a cake with robotic parts that can be eaten? One such cake was recently created - and it wasn't done on a whim, for social media clout or to simply make a unique dessert. Rather, this RoboCake is being considered as an indicator of the progress made in edible technology that expands the horizons of both gastronomy and robotic science. Created as part of the EU-funded RoboFood project, this "robotic wedding cake" was the product of a collaboration between scientists from the Swiss Federal Institute of Technology Lausanne (EPFL) and the Italian Institute of Technology (IIT), as well as culinary experts and confectioners from the Lausanne School of Hotel Management (EHL). It was showcased at the Expo 2025 in Osaka, which took place in mid-April. One of the most innovative parts of this "RoboCake" was edible rechargeable batteries, which were made of vitamin B2, quercetin, activated carbon and chocolate. They were used to light the LED candles on the cake. IIT researchers, coordinated by Mario Caironi, were the ones who crafted these batteries - reportedly the first of their kind in the world. According to Valerio Galli, a PhD student at IIT, "The first flavour you get when you eat them is dark chocolate, followed by a surprising tangy kick, due to the edible electrolyte inside, which lasts a few seconds." As per the researchers, the batteries could potentially help combat the burgeoning problem of electronic waste. Edible robotic teddy bears Photo Credit: EPFL/Jamani Caillet Another highlight of the cake were two completely edible robotic teddy bears, which were placed on the topmost tier. Scientists at the Laboratory of Intelligent Systems (LIS) at EPFL used gelatin, syrup and colorants to make the bears, which can also move because of a specific mechanism. Bokeon Kwak, a researcher at LIS, explains, "They are animated by an internal pneumatic system: when air is injected through dedicated pathways, their heads and arms move." These bears were said to taste like soft, sweet pomegranate gummies. The engineers worked with the chefs from EHL to ensure these robotics parts were safe to eat and tasted good. Dario Floreano, head of the LIS, noted, "Robotics and food are two separate worlds. However, merging them offers many advantages, particularly in terms of limiting electronic waste and food waste... Edible robots could be used to deliver food to endangered areas, to deliver medicines in innovative ways to people who have difficulty swallowing or to animals, or even to monitor food and its freshness using sensors that can be eaten."
Business Mayor
27-04-2025
- Business
- Business Mayor
What Happens When Legal Minds Meet Machine Intelligence?
Ex Nunc Intelligence is a Swiss-based legal technology company – the 1st LegalTech from EPFL – that empowers legal professionals with advanced AI tools to navigate complexity and redefine how they access, manage and leverage legal knowledge. The founding team — Zoé, Kyriaki, and Thomas — brings a unique blend of legal and tech expertise. Before starting the venture, Zoé (tax specialist) and Kyriaki (lawyer) were working together and experienced firsthand the recurring inefficiencies in legal research and knowledge management. Thomas, with his deep expertise in AI and data science, joined to help turn that insight into a scalable, intelligent platform. The idea for Ex Nunc was born out of real-world frustrations and a shared ambition to transform legal knowledge into a strategic asset. Our mission is to redefine the way legal professionals access, manage, and leverage legal knowledge. By combining cutting-edge AI with a structured legal database, we provide a strategic assistant that helps users reduce time spent on complex legal tasks from hours to minutes. Our long-term impact lies in our ability to turn fragmented legal data into a reliable and scalable decision-support infrastructure. Silex, our platform, acts as a legal co-pilot. It digests vast volumes of legal texts, jurisprudence and other legal documents to provide clear, context-aware answers and drafting assistance. This helps legal departments and law firms reduce uncertainty by basing their planning on complete, up-to-date, and verifiable legal information, significantly mitigating risks linked to gaps or inconsistencies in knowledge. Our primary users are legal professionals: law firms, corporate legal departments, insurances with legal protections, financial and public institutions. Each of them shares a core need: conducting legal research efficiently. That's where Silex comes in — by offering direct, structured, and intelligent access to the legal information they need. But beyond research, Silex also supports drafting, providing users with substantive-level control, not just formal structuring. This is made possible thanks to our proprietary legal database and the way our AI models are trained to master legal knowledge, reasoning, and methodology. We co-develop features through our Legal Lab to ensure that Silex integrates seamlessly into their specific workflows, terminology, and jurisdictional requirements. Our platform continuously processes and indexes legal updates, regulatory changes, and new case law. Through our proprietary pipeline, these signals are translated into alerts, summaries, and suggested actions. By structuring the data and applying intelligent filters, we help users identify what truly matters early on and what concrete actions to take. We do not position ourselves as a risk analytics or scenario planning company. What sets us apart is our focus on equipping legal professionals with structured, pre-processed, and context-aware legal information, enabling them to make faster, more informed decisions. We may suggest relevant directions or highlight critical insights, but the strategic choices always remain in the hands of the professional. Rather than replacing legal judgment, Silex enhances it — by surfacing the right information at the right time, grounded in solid legal reasoning and methodology. One of the most striking reactions we often witness during product demos is when we let legal professionals try Silex themselves. They enter a legal question — sometimes quite complex — and within seconds, they get a clear, structured, and sourced answer. At that moment, you can see the surprise on their faces, and they react spontaneously with genuine amazement — often explaining that this exact research took them hours, sometimes even days, to complete ! It's a powerful moment that speaks for itself. It's not just about saving time — it's about experiencing a true paradigm shift in how legal knowledge is accessed and used. We've seen this reaction consistently across law firms, corporate legal teams, and public institutions, confirming that Silex fundamentally transforms the legal research workflow. One of the biggest challenges we faced was harmonizing the chaos of legal data — ensuring consistent quality despite the vast disparity in formats, sources, and structures. Bringing together the various sources of law into a unified, reliable database required both hard work and smart engineering. Another major challenge was making a search engine work at scale: retrieving the right information quickly and accurately across millions of data points. This involved building robust indexing pipelines, developing custom preprocessing tools, and continuously optimizing relevance and performance. Ultimately, we overcame these hurdles through a combination of deep legal expertise, strong collaboration with our users, and a relentless focus on precision, speed, and scalability. Read More QC grants P6M to 6 local tech startups - The Manila Times Our next steps include expanding across Switzerland — from French-speaking regions to German- and Italian-speaking ones. We also plan to enter in neighbourouding countries and in the EU as the second step. In terms of product we have a super exciting roadmap but we are very discrete about what we're bulinding until it's out there. Partnerships are an important pillar of our growth strategy. We collaborate not only with public institutions, but also with private organizations. Several exciting partnerships are currently being finalized, and we look forward to sharing more details soon. We are always open to new collaborations that align with our vision. What three pieces of advice would you give to other founders building tech-driven solutions in highly complex domains? Co-create with your users from day one — complexity requires alignment with real practices. Invest early in data architecture — your models are only as good as your underlying data. Don't wait for perfection — deploy fast, learn continuously, and iterate openly. Picture: Ex Nunc Intelligence founder team picture Zoé Berry and Kyriaki Bongard Picture@ Marie-Lou Dumauthioz/Tamedia Thank you Zoé Berry and Kyriaki Bongard for the Interview Statements of the author and the interviewee do not necessarily represent the editors and the publisher opinion again.
Yahoo
15-04-2025
- Science
- Yahoo
Scientists issue urgent warning after discovering major threat in the Arctic: 'We need to protect them'
By now, global warming is a common talking point, but what does it tangibly mean for us when temperatures rise and glaciers start to melt? A new study led by the Swiss Federal Technology Institute of Lausanne (EPFL) revealed what glacier melt could bring about on our planet by the end of the century. EPFL scientists established a link between the microbial ecosystems on glaciers and the nutrients in the water of glacier-fed streams. The researchers believe that without the extreme cold of glacial environments, the bacteria found in those regions will likely die out. "These microorganisms … purify the water, recycle nutrients and regulate important cycles like those involving nitrogen and carbon," said Tom Battin, who co-authored the study. "Ultimately, they support the food chain and are crucial to many environmental equilibria." Since glaciers hold almost 70% of the world's freshwater, according to the National Snow and Ice Data Center, the loss of these microbial systems would trickle down to the human communities sustained by glacier-fed water. You don't need to live near the Arctic to feel the effects of glacier melt. In addition to threatening helpful, water-purifying bacteria, melting glaciers risk releasing other frozen microbes into meltwater, some of which may result in disease for local populations. For instance, researchers at the Chinese Academy of Sciences in 2022 encountered several unfamiliar viruses trapped in a 15,000-year-old iceberg. When ice melts, the water runoff may continue to carry potentially infectious microbes. The rising sea levels worldwide also jeopardize coastal homes, buildings, and roads. While changing global temperatures may not directly cause extreme weather events such as hurricanes and earthquakes, the higher water levels exacerbate their damage. "Higher background water levels mean that deadly and destructive storm surges, such as those associated with Hurricane Katrina, 'Superstorm' Sandy, and Hurricane Michael, push farther inland than they once did," wrote Rebecca Lindsey of There's no spot treatment for glaciers, but countries and organizations worldwide are working to cut down on greenhouse gas emissions and implement stronger climate policies to mitigate the effects of climate change. Several businesses are holding themselves to a clean-energy mission, foregrounding renewable energy and water conservation. Do you think America does a good job of protecting its natural beauty? Definitely Only in some areas No way I'm not sure Click your choice to see results and speak your mind. You can join the effort against climate change from your own home. Opting for EVs or public transportation, repurposing your waste to avoid landfill buildup, choosing energy-efficient appliances, and installing solar panels can all help you lower your contributions to climate change. "If we don't want to lose these irreplaceable habitats as glaciers melt," said Massimo Bourquin, lead author of the EPFL study, "we need to protect them." Join our free newsletter for good news and useful tips, and don't miss this cool list of easy ways to help yourself while helping the planet.
