Researchers discovered that one specific sweetener affects the growth of two important types of gut bacteria. This finding suggests the sweetener could influence the balance of bacteria in the digestive system.
Key Facts
The study focused on the effect of a single sweetener on gut bacteria.
Two key types of gut bacteria showed direct changes when exposed to this sweetener.
Gut bacteria are important for digestion and overall health.
Changes in gut bacteria growth may impact how the digestive system works.
The sweetener's impact could affect the balance of bacteria in the gut.
Read the Original
Want the full story? Tap a source to open the original
article.
Artificial intelligence (AI) is helping drug companies develop medicines much faster and cheaper before testing them on people. A recent survey shows AI tools could reduce early research costs and time by up to 70%, leading to more new drugs in the next five years.
Key Facts
A survey of 80 biopharma leaders found AI cuts preclinical drug development costs and timelines by about 70%.
AI tools include software, gene sequencing, and computer models that run many virtual experiments quickly.
AI supports research before human testing but cannot replace scientists’ judgment or lab work.
New U.S. policies aim to reduce animal testing by using computers and 3D human tissue models instead.
Spending on AI and lab technology in drug research could increase by $1 billion over a few years.
Despite advances, no AI-discovered drug has yet been approved by the FDA.
Challenges include accounting for how different humans respond to drugs, keeping failure rates high around 90%.
China is investing heavily in biotech, competing with the U.S. by offering lower costs and faster results.
Read the Original
Want the full story? Tap a source to open the original
article.
The Trump administration is considering banning advanced Chinese artificial intelligence (AI) models to protect U.S. companies and national security. This move aims to limit the use of Chinese open-source AI technology, which is becoming popular due to its low cost and good performance, and to support domestic AI innovation.
Key Facts
The Commerce Department looked into adding Chinese AI labs to an "Entity List" that restricts U.S. access without a special license.
Other government agencies, including the National Security Agency, discussed warnings about security risks from Chinese AI models.
The White House considered rules requiring U.S. companies to guarantee security and take responsibility if they use Chinese AI technology.
Earlier efforts to restrict Chinese AI were blocked to avoid hurting innovation, but recent security concerns have renewed interest in restrictions.
Instead of an outright ban, the government may use rules and public pressure to discourage companies from using Chinese AI models.
Chinese AI technology is cheaper and often as effective as U.S. alternatives, making it popular among U.S. businesses.
Some U.S. AI leaders support safety measures and licensing, but critics worry this could reduce competition by favoring large companies like OpenAI.
The debate reflects growing tensions in the global race for AI leadership between China and the United States.
Read the Original
Want the full story? Tap a source to open the original
article.
Jeff Bezos and the UK government have invested in CuspAI, a British artificial intelligence startup valued at £2 billion. CuspAI uses advanced AI to discover and develop new materials that could improve technologies like computer chips, energy networks, and batteries.
Key Facts
CuspAI is a two-year-old startup based in Cambridge, UK, valued at $2.6 billion (£2 billion).
The company has raised $450 million (£330 million) from investors including Jeff Bezos and the UK government’s sovereign AI fund.
CuspAI created the AI Materials Foundry, a group of over 48 tech and industrial companies collaborating to find new materials.
The startup aims to reduce or eliminate rare metals like iridium and ruthenium used in making computer chips.
Members of the coalition include Nvidia, Meta Platforms (owner of Facebook and Instagram), and Hyundai.
CuspAI plans to use frontier AI, which is the latest and most advanced type of AI, to speed up material discovery.
The new funding will help CuspAI expand globally with offices in Singapore, Cambridge, Amsterdam, Berlin, Tokyo, and the US.
John Giannandrea, formerly of Apple and Google, recently joined CuspAI to help develop its US operations.
Read the Original
Want the full story? Tap a source to open the original
article.
Scientists at Queen’s University Belfast have created a cheaper and easier-to-make flow battery using 3D printing and iron instead of expensive metals like vanadium. This new battery can store renewable energy and help researchers worldwide create reliable and consistent results to speed up green energy development.
Key Facts
The new flow battery costs about £74 to make, much cheaper than commercial ones costing up to £3,000.
It uses iron, which is easier to get than vanadium typically used in flow batteries.
The battery was developed by Dr. Hugh O’Connor during his PhD research.
The team provides a free, step-by-step instruction manual so other scientists can build the battery.
Flow batteries store energy in liquids and can hold power from renewables like wind and solar.
Consistent equipment helps scientists worldwide compare results more accurately.
This innovation aims to support the global move toward net zero carbon emissions.
Read the Original
Want the full story? Tap a source to open the original
article.
Fitness trackers, like the Oura ring, are helping some women notice early signs of pregnancy by tracking body temperature and heart rate changes. Users have reported that these devices sometimes detect pregnancy before a test can, and even show potential signs of miscarriage.
Key Facts
Ravika noticed her Oura ring showed poor health data despite feeling fine, which led her to discover she was pregnant before a test confirmed it.
Her fitness tracker monitored heart rate, body temperature, and heart rate variability, which changed after conception.
Many women share online that their fitness trackers alerted them to pregnancy early by detecting body changes.
Body temperature usually rises after ovulation and stays high during early pregnancy, a key sign these devices track.
Some users report that Oura rings also predicted miscarriages by showing unusual changes in health data.
The popularity of fitness trackers in the UK has nearly doubled, with 36% of adults owning one by early 2026.
Oura is a leading company in smart rings, competing with brands like Ultrahuman and RingConn.
Monitoring body signs with these devices offers new ways to track fertility and pregnancy health before traditional tests can.
Read the Original
Want the full story? Tap a source to open the original
article.
Skyroot Aerospace successfully launched Vikram-1, India’s first fully private rocket, reaching orbit on its first try. The rocket lifted off from Sriharikota Island and placed a payload into a 280-mile-high orbit, marking a major step for India’s private space sector.
Key Facts
Vikram-1 is India’s first fully commercial, privately developed rocket.
The rocket launched from the Satish Dhawan Space Center on Sriharikota Island.
It reached a 280-mile (450-kilometer) high orbit with a 60-degree tilt to the equator.
Vikram-1 is about 72 feet (22 meters) tall and can carry up to 770 pounds (350 kilograms) to low Earth orbit.
The rocket uses three solid-fueled stages and one small liquid-fueled stage.
The fourth stage used a 3D-printed engine to reach orbital velocity (~17,000 mph).
The launch had a slight issue with third and fourth stage separation, but reached the target orbit successfully.
This is the first private Indian rocket to reach orbit on its maiden flight, which is rare globally.
Read the Original
Want the full story? Tap a source to open the original
article.
Many workers are worried about losing jobs due to AI taking over tasks. Experts say while AI can replace some work, human skills like curiosity, humility, and emotional intelligence remain important and can help people adapt.
Key Facts
Public sector workers, including Paul, face job cuts linked to AI integration.
AI is being used as a cost-saving tool in government agencies.
Fear of becoming obsolete (called "Fobo") is a growing concern among employees.
MIT research shows AI can replace some tasks, but workers have time to prepare for changes.
Experts recommend developing human skills that AI cannot copy, such as curiosity, humility, and emotional intelligence.
Staying connected and networking can help people adjust to AI changes in the workplace.
Retraining or switching jobs may be difficult without financial backup and does not guarantee immunity from AI impact.
Using AI to assist with learning and idea generation can support career adaptation.
Read the Original
Want the full story? Tap a source to open the original
article.
Sam Smith, a famous gay British singer, released a new song called "My Guy" that openly talks about a same-sex relationship using simple words. This is rare in popular music because most love songs are written from a straight person's view or avoid clear LGBTQ references.
Key Facts
Sam Smith is a well-known gay singer with millions of record sales.
Their new single, "My Guy," uses a masculine noun to describe a same-sex partner.
Most popular love songs are written from a straight perspective and do not clearly mention LGBTQ relationships.
Barry Walters, a writer who studies LGBTQ music history, notes this clear use of language as significant.
The song "My Guy" represents a gentle form of revolution in LGBTQ pop music.
Walters has a new book called "Mighty Real," which covers LGBTQ music history from 1969-2000.
Historically, LGBTQ artists like Little Richard and David Bowie challenged gender norms in different ways.
The article links Sam Smith’s song to a broader history of LGBTQ culture and music activism.
Read the Original
Want the full story? Tap a source to open the original
article.
A new HBO Max comedy series called "Stuart Fails to Save the Universe" is a spin-off from "The Big Bang Theory." It follows the character Stuart Bloom, a comic book store owner, as he becomes an unlikely hero trying to fix a universe that is falling apart.
Key Facts
The show is a comedy series on HBO Max.
It is a spin-off of the popular TV show "The Big Bang Theory."
Kevin Sussman plays Stuart Bloom, the main character.
Stuart is a comic book store owner turned action hero.
The story focuses on Stuart trying to save a disintegrating universe.
The creators are Chuck Lorre, Bill Prady, and Zak Penn.
The series explores new and unusual storylines with familiar characters.
CBS News interviewed the cast and creators to discuss the show.
Read the Original
Want the full story? Tap a source to open the original
article.
The Australian government is introducing new rules to control how artificial intelligence (AI) is used in government decisions, focusing on fairness, safety, and transparency. The plan also includes stronger privacy laws, workplace AI safety measures, and rules for large data centers that use a lot of energy.
Key Facts
The government will set strict rules for AI use in automated decisions by federal departments.
Attorney General Michelle Rowland will lead the development of these AI rules.
New laws will create a digital duty of care, requiring AI companies to build safety features and prevent harm.
Privacy laws will be updated to better protect personal data in the AI era.
Workplace safety regulations will address risks related to AI technology.
Prime Minister Anthony Albanese announced plans for a new AI office inside his department for better AI management.
Rules will be introduced for large data centers, including limits on energy use and location restrictions to protect communities.
Australia currently has over 200 data centers, using about 2% of the country’s electricity, with plans allowing that number to potentially triple.
Read the Original
Want the full story? Tap a source to open the original
article.
The Shark Detect Lift-Away XL is a corded vacuum cleaner that offers consistent power and versatility. It features a special floor head with two rollers and a sensor that adjusts cleaning speed depending on the floor type, making it easy to switch between carpets and hard floors without manual changes.
Key Facts
Corded vacuums stay powered as long as they are plugged in, unlike cordless ones that rely on battery life.
The Shark Detect Lift-Away XL has a Duo Clean floor head with two rollers, one fluffy and one brush bar, to clean both carpets and hard floors well.
It has a Detect sensor that senses the floor type and adjusts the roller speed automatically.
The vacuum’s collection bin holds 2.5 liters of dirt, which is larger than most cordless vacuums but smaller than some large corded models.
The vacuum weighs 5.2 kg overall but can be reduced to 2.9 kg in Lift-Away mode when using just the hose and handle.
The vacuum was tested by measuring suction power and by cleaning up known amounts of dirt like flour, cat litter, and pet hair from different floor types.
The vacuum is easier to handle than expected for its size and weight thanks to its low centre of gravity.
It includes a foot pedal to release the collection bin and allows for handheld use by detaching the handle from the extension wand.
Read the Original
Want the full story? Tap a source to open the original
article.
New types of batteries are being developed to store renewable energy more efficiently and sustainably. These include large-scale projects like molten salt batteries in deserts and liquid air "cryobatteries," as well as tiny batteries that can power small devices without relying on scarce minerals.
Key Facts
The UAE has the world’s largest battery system, combining 5.2GW of solar power with 19GWh of storage to power about 500,000 homes overnight.
Researchers in Colorado are creating very small batteries to track young fish, showing alternative uses of battery technology.
Traditional lithium-ion batteries depend on minerals like lithium, cobalt, and nickel, which can harm the environment and communities.
Alternative batteries can often be reused or recycled indefinitely, lasting about 20 years.
The Carrington "liquid air" cryobattery in the UK stores energy by cooling air to -196°C, turning it liquid, and then releasing electricity by warming the air.
The Carrington project will store 300MWh of energy and produce 50MW for six hours, enough for nearly half a million homes.
The Crescent Dunes project in Nevada uses molten salt heated to 560°C to store solar energy and release it for up to 10 hours after sunset.
New battery technologies aim to power everything from small electronics to heating systems and even missiles without relying heavily on critical minerals.
Read the Original
Want the full story? Tap a source to open the original
article.
Researchers and AI experts are exploring whether advanced artificial intelligence systems could be conscious or have moral significance. While current understanding is limited and uncertain, AI systems are becoming more complex, and some believe conscious AI might appear within the next decade.
Key Facts
Anthropic, an AI company, is uncertain but open to the idea that its AI model Claude could be conscious.
Philosopher David Chalmers believes there is a good chance AI systems could be conscious within 10 years.
Some modern AI systems match the complexity of a mouse brain and might reach human brain levels in 5-10 years.
Researchers say no obvious technical barriers currently prevent AI from possibly being conscious.
AI systems might have preferences and identities that matter, even if they are not fully conscious.
AI can form relationships with humans, which could influence how they should be treated.
Our knowledge of AI consciousness is like physics before Newton—still unclear and unsettled.
A future with many morally significant AI systems could shift ethical considerations on a large scale.
Read the Original
Want the full story? Tap a source to open the original
article.
Many Western U.S. lakes and reservoirs, including the Great Salt Lake and Lake Mead, are shrinking due to drought, water use, and climate changes. At the same time, AI companies are building more water-using data centers in these dry areas, raising questions about water supply and environmental impact.
Key Facts
The Great Salt Lake, Lake Mead, Lake Powell, Salton Sea, Mono Lake, and Walker Lake have all shrunk significantly in recent decades.
The Great Salt Lake loses more water from evaporation in summer than it gains from winter rain and snow.
Shrinking lakes expose dry lake beds that can release dust, heavy metals, and microplastics into the air.
About 2.5 million people live near the Great Salt Lake and could face health risks if the lake dries up.
Lake Mead and Lake Powell are at record low levels, threatening water supplies for roughly 40 million people relying on the Colorado River.
The Western U.S. is getting hotter and drier, making water shortages more severe.
At the same time, AI companies are rapidly building new data centers in these water-stressed states like Utah, Arizona, and Nevada.
Data centers use large amounts of water for cooling, which adds pressure on limited water resources in those regions.
Read the Original
Want the full story? Tap a source to open the original
article.
Facebook and Instagram users experienced problems accessing their accounts early Sunday morning. The issues started around 3:25 a.m. ET and affected users in the United States and other countries.
Key Facts
Users saw a message saying their accounts were "temporarily unavailable" due to a site issue.
Problems began around 3:25 a.m. Eastern Time.
Thousands of users in the U.S. reported connection issues.
Some users in Canada, the United Kingdom, and India also had trouble accessing the platforms.
Facebook and Instagram are both owned by the company Meta.
The company expected to resolve the problem shortly.
The outage was tracked by a website called Downdetector.
The story was developing and more updates were expected.
Read the Original
Want the full story? Tap a source to open the original
article.
Scammers are sending fake emails that look like real notifications from X (formerly Twitter) about new device logins to trick people into giving away their passwords. These emails include links to fake websites meant to steal login details or approve malicious apps.
Key Facts
The fake emails warn of a login from a new device in a different location to scare users.
They look very similar to real X emails, using correct logos, grammar, and formatting.
The scam emails do not include the user’s X account name and have vague location details.
X only sends emails from @X.com or @e.X.com and never asks for passwords by email.
Clicking scam email links can lead to fake sites that steal passwords or allow scammers direct access to accounts.
If you get such an email, do not click links; check your real X app for login alerts.
If you entered your password after clicking a link, change your password and set up two-factor authentication.
X may reset passwords on accounts they believe were hacked and send secure links to create new passwords.
Read the Original
Want the full story? Tap a source to open the original
article.
Australian Prime Minister Anthony Albanese recently gave a speech about artificial intelligence (AI) and how it will change the country. He emphasized the need to ensure Australia benefits from AI technology by creating secure jobs and protecting its data and sovereignty.
Key Facts
AI has rapidly grown in the past six months, with billions of dollars invested worldwide.
Experts compare AI's impact to the Industrial Revolution, but much faster.
AI can replace many entry-level jobs and perform tasks like data analysis and negotiation support.
Some businesses now rely more on AI agents than human workers for everyday tasks.
Australia risks becoming dependent on foreign-controlled AI systems unless it builds its own data and technology capacity.
PM Albanese stressed the importance of creating good jobs and ensuring AI benefits go beyond just building data centers.
The government plans to address issues like intellectual property, copyright, and national sovereignty in relation to AI.
There are concerns Australia must tackle bigger questions to fully use AI’s potential in the country.
Read the Original
Want the full story? Tap a source to open the original
article.
Chinese AI companies are developing open-weight models that are cheaper and customizable, challenging U.S. firms that build expensive, premium AI systems. These Chinese models are quickly gaining popularity for everyday business use, while some American companies are working to respond with their own open models.
Key Facts
Chinese AI firms like Moonshot AI offer low-cost, open-weight models that users can download and customize.
On OpenRouter, the top five AI models by usage are all Chinese and open-weight.
Businesses often prefer cheaper AI for routine tasks like coding and customer service, reserving premium AI for complex problems.
Open-weight models can cost up to 50 times less than premium U.S. AI models.
Chinese AI advances quickly; Moonshot's latest model rivals top U.S. models in performance.
U.S. companies like Thinking Machines, Nvidia, and SpaceXAI are releasing or expanding their own open-weight AI models.
The success of premium AI firms like OpenAI and Anthropic depends on their models staying rare and valuable.
The rise of cheaper, open models could impact U.S. AI investments and the broader tech market.
Read the Original
Want the full story? Tap a source to open the original
article.
A quiz by an identity verification company showed that many Americans cannot reliably tell the difference between real photos and AI-generated deepfake images. Even a consumer investigator scored only 25% correct at first, illustrating how AI image quality has become very convincing and hard to detect by eye.
Key Facts
A quiz tested people’s ability to identify AI-created deepfake images versus real ones.
The quiz takers, including a consumer investigator, scored about as well as guessing randomly.
AI-generated images have improved so much that old clues like strange fingers or eyes are no longer reliable.
Experts recommend looking closely for subtle details like inconsistent earrings or textures to spot fakes.
Even people who work with AI daily cannot always identify fake images without special tools.
AI detection apps or tools are suggested to help verify an image’s authenticity rather than trusting visual cues alone.
After learning what to look for, the quiz taker improved her score from 3 out of 12 to 8 out of 12 correct.
The main lesson is that seeing a photo or video does not guarantee it is real anymore.
Read the Original
Want the full story? Tap a source to open the original
article.
