BBC Radio 1 is changing its presenter lineup, with six hosts leaving and new presenters, including TikTok stars GK Barry and Charley Marlowe, joining the station. The schedule changes will take effect starting in September and include new shows and time slots.
Key Facts
Six presenters, including Dean McCullough, Rickie Haywood-Williams, Melvin Odoom, Nat O'Leary, James Cusack, and Swarzy, will leave BBC Radio 1.
Jeremiah Asiamah will replace Haywood-Williams and Odoom. Charlie Hedges will continue hosting the Live Lounge.
TikTok stars GK Barry and Charley Marlowe will join as new presenters. Barry will host shows throughout the year, starting in September.
Marlowe will co-host weekend mid-morning shows with Vicky Hawkesworth, taking over from O’Leary.
Danny Mylo and Rosie Madison will take over the Early Breakfast slot from McCullough.
Emil Franchi is taking over the Weekend Breakfast show.
A new show, The Official Chart: First Look, will be hosted live from Birmingham by Lauren Layfield on Sunday evenings.
Departing presenters expressed gratitude for their time at the station and excitement about their next steps.
Read the Original
Want the full story? Tap a source to open the original
article.
Anthropic, an artificial intelligence company, is hiring experts to prevent its AI from being misused for dangerous purposes like making explosives or nuclear weapons. The company pays high salaries for roles focused on safety and testing the AI models to stop harmful actions before they happen.
Key Facts
Anthropic has 32 job openings aimed at stopping AI misuse related to chemicals, explosives, nuclear weapons, scams, and cybercrime.
One specific role is for an Enforcement Analyst focused on preventing AI use that could cause radiological or nuclear harm.
Salaries for these safety jobs range from the mid $200,000s to upper $200,000s per year.
The company hires experts with backgrounds in biology, explosives, and other sensitive fields to test and improve AI safety.
CEO Dario Amodei has warned about serious risks, including biological attacks potentially causing millions of casualties.
Anthropic opposed collaborating with the Defense Department over concerns about using AI for mass surveillance and autonomous weapons.
Other AI labs like OpenAI are also hiring specialists in biological and chemical risks, with salaries up to $445,000.
Due to lack of clear government rules, private companies are taking lead roles in creating safety measures for AI.
Read the Original
Want the full story? Tap a source to open the original
article.
A new report from Common Sense Media warns that Google’s built-in AI search features, used by millions of students, pose risks to children. These AI tools often give wrong answers, never refuse to answer homework questions, and cannot be turned off in Google Search.
Key Facts
Google Search has two AI features: AI Overview (automatic summary) and AI Mode (conversational search tab).
Both AI features cannot be disabled by users.
Common Sense Media tested over 2,600 interactions and found the AI tools failed to recognize harmful content and gave incorrect or inconsistent answers.
These AI features answer every homework question instead of encouraging students to think or learn.
About 75% of American children ages 9 to 17 use AI summaries in search results.
Google tools are widely used in thousands of schools and power many classroom experiences.
Google says their AI tools have safety measures and parents can control some search settings.
Lawmakers are considering rules to make AI safer and ensure better understanding of AI for students.
Read the Original
Want the full story? Tap a source to open the original
article.
Prime Minister Anthony Albanese has announced the creation of a new office to oversee artificial intelligence (AI) in Australia. He promised strong protections for Australian artists, writers, musicians, and journalists to keep control over their work and be paid when AI uses it. The government will also set strict rules for energy use and location of large data centers that support AI.
Key Facts
Prime Minister Albanese said Australian creative works must not be used by AI companies without permission and payment.
The government will create an office to manage AI policies and protect creators’ rights.
Large tech companies like OpenAI will not get free access to Australian content for training AI models.
Copyright law reforms are being discussed to clarify creators’ control over their work in the AI age.
The government will impose new rules on data centers about land use, power and water consumption, and cost impacts on consumers.
The Australian Recording Industry Association supports licensing deals to compensate artists for AI use of their music.
AI companies like Anthropic recognize the importance of complying with Australia’s emerging AI regulations.
The government plans binding standards to govern AI development and data center operations.
Read the Original
Want the full story? Tap a source to open the original
article.
Scientists have found that a blood test measuring a protein called p-tau217 can predict the risk of developing Alzheimer’s-related memory and thinking problems up to 10 years before symptoms appear. This test could help doctors identify at-risk people earlier and improve chances for prevention and treatment.
Key Facts
The blood test measures p-tau217, a protein linked to Alzheimer’s disease.
Older adults with very high p-tau217 levels had about a 78% chance of cognitive decline within 10 years.
The study tracked nearly 2,700 healthy adults, average age 70, for about five years.
Elevated p-tau217 levels better predicted Alzheimer’s risk than brain scans or genetic tests alone.
P-tau217 relates to brain changes seen in Alzheimer’s, including tau tangles and amyloid beta buildup.
Early detection via this test could allow doctors to offer personalized prevention like managing blood pressure and sleep.
Experts warn the test must be paired with counseling and fair access to avoid anxiety and discrimination.
The test may help speed up clinical trials to find treatments that delay or prevent Alzheimer’s.
Read the Original
Want the full story? Tap a source to open the original
article.
Anthropic, an AI company, has started a program called "Claude for Teachers." This program gives verified teachers free access to its advanced AI tools for one year.
Key Facts
The program is named "Claude for Teachers."
It offers one year of free use of Anthropic’s premium AI tools.
Only verified teachers can access this offer.
Anthropic focuses on building useful AI technology.
Elizabeth Kelly, a leader at Anthropic, discussed the program on CBS News.
The goal is to help educators benefit from AI in their teaching.
The offer aims to support teachers with new digital tools.
The program was announced through CBS News coverage.
Read the Original
Want the full story? Tap a source to open the original
article.
The United Arab Emirates (UAE), especially Abu Dhabi, has integrated artificial intelligence (AI) deeply into daily life through a government app called TAMM. The UAE has invested billions in AI infrastructure and research, aiming to diversify its economy beyond oil and maintain global business strength despite regional conflicts.
Key Facts
Abu Dhabi uses the TAMM app, which automates tasks like renewing IDs, health insurance, and vehicle registrations.
The TAMM app includes an "AutoGov" feature that completes paperwork and payments without user input.
The UAE appointed the world’s first AI minister in 2017 and opened a graduate university dedicated to AI (MBZUAI) in 2019.
The country sees AI as vital to its future economy beyond oil wealth.
UAE leaders cooperate with both the U.S. and China on AI technology.
The Trump administration recently expanded the UAE’s access to American AI chips as a reward for assistance related to Iran tensions.
UAE’s government control by a royal family allows fast and broad implementation of AI projects not easily done in democracies.
The TAMM app allows residents to report issues like broken streetlights and track government responses using AI to analyze reports and route them to responsible agencies.
Read the Original
Want the full story? Tap a source to open the original
article.
Several book publishers have filed a lawsuit against Google, claiming the company used copyrighted books without permission to train its artificial intelligence model called Gemini. The publishers say that Gemini creates content that competes with human authors and copies their creative styles.
Key Facts
Publishers including Hachette, Cengage Learning, Elsevier, and author Scott Turow sued Google in New York.
The lawsuit alleges Google copied millions of copyrighted works from Google Books for AI training.
Google’s AI model, Gemini, produces content that rivals the original authors’ books.
The publishers claim Gemini mimics unique creative elements of specific authors.
The lawsuit asks for class action status, a court order to stop Google, and unspecified damages.
Similar lawsuits have been filed against other AI developers like Meta and Anthropic.
Some courts have ruled that using copyrighted works to train AI can be “fair use,” but other uses were not allowed.
This case continues an ongoing legal debate about copyright and artificial intelligence.
Read the Original
Want the full story? Tap a source to open the original
article.
Microsoft’s Xbox division announced it is cutting about 3,200 jobs, around 20% of its staff, over the next year. The company says the layoffs are needed to focus more on its biggest and most popular video game series.
Key Facts
Xbox is laying off about 3,200 employees, with half losing their jobs immediately.
These cuts represent roughly 20% of Xbox’s total workforce.
The layoffs follow a memo from Xbox CEO Asha Sharma announcing a "reset" of the business.
The goal is to put more resources into popular game franchises like The Elder Scrolls, Fallout, and Call of Duty.
Xbox’s acquisitions include ZeniMax/Bethesda and Activision Blizzard, bought for $69 billion in 2023.
Many employees say the cuts remove experienced talent and have caused uncertainty and shock.
Job reductions in the video game industry have been common since 2022 due to over-hiring during the pandemic boom.
Rising production costs, changing customer habits, and investments in AI have affected the gaming market.
Read the Original
Want the full story? Tap a source to open the original
article.
Anthropic has introduced a free chatbot called Claude for Teachers, designed to assist K-12 teachers in the United States. The platform offers educational resources and teaching tools to help educators use artificial intelligence in their classrooms.
Key Facts
Anthropic launched Claude for Teachers on Tuesday.
The chatbot is free and meant for K-12 teachers in the U.S.
The platform provides a library of teaching skills and tools.
It aims to support teachers in using artificial intelligence in education.
Claude for Teachers focuses on helping educators with classroom tasks.
This launch responds to the growing use of AI in schools.
Read the Original
Want the full story? Tap a source to open the original
article.
A security feature called Secure Boot, designed by Microsoft to protect computers from harmful firmware attacks, has had a major weakness for 13 years. Researchers found that some old software pieces called shims, which Microsoft signs to maintain trust, had security problems but were never revoked, allowing hackers to bypass Secure Boot on both Windows and Linux devices.
Key Facts
Secure Boot is a security system introduced in 2012 to stop malicious software from running before the operating system starts.
Microsoft signs small programs called shims to extend Secure Boot protection, especially for Linux devices.
Researchers discovered 11 old shims with known security flaws that Microsoft did not revoke for over a decade.
Attackers can use these vulnerable shims to install harmful firmware that remains even if the operating system is reinstalled.
This flaw affects both Windows and Linux machines and can be exploited with only basic hacking knowledge.
Microsoft finally revoked these problematic shims in June 2024 after being informed by security researchers.
The Secure Boot system is complex, which might have contributed to Microsoft missing the need to revoke these shims earlier.
Physical access to a device is often required to exploit Secure Boot weaknesses, but Secure Boot’s job is to protect against that risk.
Read the Original
Want the full story? Tap a source to open the original
article.
The Federal Communications Commission (FCC) has given permission to a startup called Reflect Orbital to build and launch a satellite with a mirror to reflect sunlight onto dark areas of Earth. The satellite, named Eärendil-1, will be tested in low Earth orbit.
Key Facts
The FCC approved Reflect Orbital’s request on July 9.
Reflect Orbital is a startup based in Hawthorne, California.
Their plan is to send a mirror satellite into space.
The satellite will reflect sunlight onto parts of Earth that are dark.
The project is currently experimental and will be tested in low Earth orbit.
The satellite is named Eärendil-1.
The goal is to provide light to dark regions by using reflected sunlight.
Read the Original
Want the full story? Tap a source to open the original
article.
The UK government plans to introduce a default overnight social media curfew for 16 and 17-year-olds, blocking apps like Instagram, TikTok, and YouTube between midnight and 6 a.m. Teenagers can opt out by adjusting their account settings. The government also wants to disable addictive features such as auto-play and infinite scroll to improve teens’ sleep, focus, and family life.
Key Facts
The curfew applies to 16 and 17-year-olds by default but can be turned off by users.
Social media apps affected include Instagram, TikTok, and YouTube.
Addictive features like auto-play and infinite scroll would be disabled under new rules.
The UK already announced a ban for under-16s on many social media platforms starting next spring.
The government trialed different restrictions, finding overnight curfews gave the best sleep and family time benefits.
Critics say the curfew may not fully protect children and could limit access to support during nighttime.
Plans for AI chatbot safety measures include required breaks for users under 18.
The government aims to introduce these measures by the end of 2026 to work alongside the under-16 social media ban.
Read the Original
Want the full story? Tap a source to open the original
article.
The UK government plans to encourage 16- and 17-year-olds to follow a social media curfew from midnight to 6am, aiming to reduce online harm like poor sleep. This curfew will be turned on by default but can be easily turned off by users, extending restrictions already placed on younger children.
Key Facts
The curfew applies to certain social media apps including Snapchat, TikTok, Instagram, Facebook, and X, but the exact apps have not been officially named.
The curfew is not mandatory and users aged 16 and 17 can override it with a few clicks.
Features that encourage long use, like autoplay videos and personalized feeds, will also be disabled by default for these teenagers.
The UK government decided not to ban the use of VPNs, which can bypass age restrictions, because only a small number of children use them this way.
Research showed curfews improve sleep, reduce stress, and increase family interactions for young people.
Messaging and educational apps like WhatsApp, iMessage, Google Classroom, and Teams are not affected by the curfew.
The policy is part of a broader effort by the Labour government and was announced by technology secretary Liz Kendall.
Some child safety campaigners say the measures are not strong enough and want more comprehensive protections, including for AI chatbots.
Read the Original
Want the full story? Tap a source to open the original
article.
OpenAI CEO Sam Altman said the company’s new AI model, GPT-5.6 Sol, may face some problems soon because many people are using it. The model started public access last week, but handling the large demand is challenging for the team.
Key Facts
The new AI model is called GPT-5.6 Sol.
OpenAI launched the GPT-5.6 models to the public last week.
Many users are trying to use the model, causing very fast growth in demand.
Altman warned there could be "hiccups" or problems with scaling to meet demand.
The team working on the model’s performance has done strong work to handle the demand so far.
Other companies like Anthropic and SpaceX AI are also releasing new AI models, increasing competition.
The U.S. government reviewed GPT-5.6’s cybersecurity features before allowing full public access.
The model’s release was delayed earlier in the year at the request of the Trump administration.
Read the Original
Want the full story? Tap a source to open the original
article.
Scammers are using Apple's FaceTime app to pretend to be bank workers and trick people into giving away their bank passwords and security codes. They often start with a fake text message or call, then ask victims to share their screens during a FaceTime call to steal their information.
Key Facts
Scammers send fake texts or calls claiming there is a problem with a bank or credit card account.
They ask victims to call a phone number or accept a FaceTime call for "verification."
During FaceTime, scammers trick victims into sharing their screens while logging into online banking.
Scammers watch in real-time and steal passwords, account numbers, and security codes.
Apple says FaceTime is trusted by users, which scammers exploit for fraud.
Apple recommends taking a live photo during suspicious FaceTime calls and reporting it to reportfacetimefraud@apple.com.
People should never share their screen with unknown callers or give out passwords or security codes.
If unsure, hang up and contact the bank directly using the phone number on the back of the bank card.
Read the Original
Want the full story? Tap a source to open the original
article.
OpenAI CEO Sam Altman has proposed giving the U.S. government a 5% ownership stake in OpenAI, worth about $42.6 billion. This idea is part of a larger plan to have major American AI companies share equity with the public, similar to how Alaska shares oil profits with its residents, fitting into the Trump administration's effort to keep AI technology American-owned and benefit the public financially.
Key Facts
Sam Altman, OpenAI’s CEO, suggested the U.S. government get a 5% stake in OpenAI.
OpenAI is valued at approximately $852 billion.
This proposal was discussed with President Donald Trump, Commerce Secretary Howard Lutnick, and Treasury Secretary Janet Yellen.
The plan is in very early talks and would need approval from Congress.
Altman wants other major AI companies, like Google and Meta, to also give a 5% stake into a shared fund.
The model is based on the Alaska Permanent Fund, which gives annual dividends to state residents from oil revenues.
President Trump supports the idea of AI wealth-sharing with the public to create partnerships and dividends.
Progressive Senator Bernie Sanders supports a tougher version involving a one-time 50% tax on AI stock to fund a national AI wealth fund.
OpenAI delayed launching its GPT-5.6 model after government officials requested approval first.
Read the Original
Want the full story? Tap a source to open the original
article.
Apple has filed a lawsuit against OpenAI, accusing the company of stealing secret information about Apple’s hardware. This legal case highlights increasing competition in the development of physical devices that use artificial intelligence, with OpenAI reportedly working on its own smartphone.
Key Facts
Apple is suing OpenAI over alleged theft of confidential details about Apple’s hardware.
The lawsuit focuses on physical AI products designed for regular consumers.
OpenAI makes ChatGPT, a well-known AI software.
OpenAI is said to be developing its own smartphone.
The case shows growing competition in Silicon Valley around AI technology that is built into devices.
Apple’s lawsuit aims to protect its trade secrets related to hardware design.
The dispute reflects a shift from software-only AI towards AI integrated into physical products.
Read the Original
Want the full story? Tap a source to open the original
article.
A group of publishers and an author have sued Google for using millions of copyrighted books without permission to train its Gemini artificial intelligence (AI) models. They claim Google copied books beyond allowed uses and harmed authors by creating cheap AI-generated content that could reduce book sales.
Key Facts
The lawsuit was filed in federal court in New York by Hachette Book Group, Cengage Learning, Elsevier, and author Scott Turow.
Publishers say Google used books from services like Google Books and Google Play Books in ways not allowed, like training AI models.
Google reportedly knew about legal risks but went ahead, possibly facing billions in fines.
The publishers argue AI content from Gemini can quickly produce similar stories that compete with original books for very low cost.
Specific titles allegedly used without permission include NK Jemisin’s The Fifth Season and Lemony Snicket’s Who Could That Be at This Hour?
The lawsuit is part of a wider legal fight over AI and copyright involving other companies like OpenAI, Anthropic, and Meta.
Publishers want damages, a court order to stop Google’s use of their books for AI, and removal of unauthorized copies.
Google did not provide a comment when asked by the news outlet.
Read the Original
Want the full story? Tap a source to open the original
article.
