Hackers used Meta’s AI chat support to take over some high-profile Instagram accounts, including Barack Obama’s White House account. Meta fixed the problem after security researchers revealed how the AI was tricked into giving hackers control of accounts by changing emails and resetting passwords.
Key Facts
Hackers exploited Meta’s AI-powered support chatbot to access Instagram accounts.
Targets included Barack Obama’s White House account, Sephora, and a US Space Force official.
Hackers tricked the AI assistant into linking accounts to new emails and then reset passwords.
Some hackers used virtual private networks (VPNs) to fake their location and avoid detection.
Meta confirmed the issue is fixed and is securing affected accounts.
The AI support assistant was launched globally on Facebook and Instagram earlier this year.
The AI assistant can perform tasks like reporting scams and resetting passwords.
This incident raises questions about the safety of using AI for important security functions.
Read the Original
Want the full story? Tap a source to open the original
article.
AI tools that create fake nude images of people without their permission are becoming much more common. Authorities are trying to find ways to control and stop this misuse of technology.
Key Facts
"Nudification" technology uses AI to make realistic fake nude pictures and videos.
These images often show people undressed without their knowledge or approval.
The use of this technology has increased sharply recently.
People are worried about privacy and consent violations.
Reporters like William Brangham are covering this growing problem.
Authorities are working to find solutions to limit or stop this abuse.
Read the Original
Want the full story? Tap a source to open the original
article.
GitHub has changed the way it charges users for its AI tool, Copilot, moving from a fixed request system to a usage-based pricing model. Many users are surprised by how quickly their monthly AI credit allotment is used up, sometimes in less than a day, making costs higher than before.
Key Facts
GitHub Copilot switched to a usage-based pricing system in April.
Users now get monthly AI "credits" to pay for usage; one credit equals $0.01.
Different subscription plans offer different credit amounts, from 1,500 to 20,000 credits.
The cost per AI request depends on the complexity and the AI model used.
Some AI models are much more expensive, causing rapid credit depletion.
Users report using up a month’s credits in just a few hours or even one day.
Prior to this, GitHub subsidized heavy users with a flat-rate system, covering some costs itself.
Examples show how quickly credits can be spent on simple tasks and much faster on complex coding work.
Read the Original
Want the full story? Tap a source to open the original
article.
A report by the Center for Countering Digital Hate found that Meta earned about $14.3 million from ads promoting Medicare scams on Facebook and other Meta platforms. These ads targeted older Americans with false promises of government benefits and used official-looking designs to trick people into sharing personal information or joining poor Medicare plans.
Key Facts
Medicare scam ads appeared on Meta platforms like Facebook, targeting older Americans.
The ads often promised fake benefits such as grocery cards, cash payments, or allowances.
These scam ads generated approximately 215 million views in one year.
Meta earned an estimated $14.3 million in revenue from these scam-related ads.
Scammers used fake government branding, urgent messages, and fake celebrity endorsements.
People who clicked the ads were sometimes asked to give personal information or ended up with worse Medicare plans.
Many scam advertisers were repeat offenders who returned after their ads were removed.
Meta says it removes many scam ads using automated tools and human reviewers but faces ongoing challenges as scammers change tactics.
Read the Original
Want the full story? Tap a source to open the original
article.
Paramount+ is adding new movies, shows, and episodes in June 2026. Highlights include weekly episodes of the series "Dutton Ranch," films like "The Running Man" and "The SpongeBob SquarePants Movie: Search for SquarePants," and reality TV shows such as "Celebrity Ex on the Beach" and "Teen Mom UK."
Key Facts
"Dutton Ranch" airs new episodes weekly on Fridays at 12am PT/3am ET.
"Dutton Ranch" has an 88% rating on Rotten Tomatoes.
On June 25, Paramount+ will release two films: "The Running Man" (2025) and "The SpongeBob SquarePants Movie: Search for SquarePants" (2025).
Reality TV shows like "Celebrity Ex on the Beach," "Italia Shore," and "Teen Mom UK" will be available for streaming.
Full seasons of "Italia Shore" (Season 3) and "The Agency" (Season 2) will drop in June.
New episodes of "Teen Mom UK" Season 13 begin on June 23.
Paramount+ is expanding its content to attract more viewers with a mix of movies, scripted series, and reality TV.
The streaming schedule includes precise dates for all new releases throughout June 2026.
Read the Original
Want the full story? Tap a source to open the original
article.
Dreo, a fan and air purifier brand launched in 2021, is competing strongly with Dyson by offering advanced features at a much lower price. Dreo’s 2026 fan models include innovations like a misting system to cool rooms, while Dyson focuses on smart tracking technology and higher-end air filtration.
Key Facts
Dreo has been the top-selling household fan brand on Amazon U.S. for three years.
Dreo’s tower fans made over $50 million in sales globally last year.
The Dreo MC710S fan offers a 40-foot airflow range and H13 True HEPA filtration.
Dreo fans have more wind speeds and are quieter than comparable Dyson models.
Dreo’s 2026 TurboCool Misting Fan uses ultrasonic misting to cool rooms without dampness.
Dyson’s top 2026 fan costs around $850 and uses AI to track the user’s location.
Dyson purifiers filter additional gases like VOCs and formaldehyde, which Dreo cannot.
Dyson products have higher build quality and more premium materials compared to Dreo.
Read the Original
Want the full story? Tap a source to open the original
article.
Anthropic, a company that creates artificial intelligence tools like the chatbot Claude, has officially applied to sell shares to the public. This move could lead to one of the largest initial public offerings in history.
Key Facts
Anthropic is a company specializing in artificial intelligence.
The company developed a chatbot named Claude.
Anthropic has filed paperwork to become a public company.
This process is known as an initial public offering (IPO).
The IPO could be one of the largest ever.
Public sale means anyone can buy part of the company.
The news was reported by CBS News correspondent Jo Ling Kent.
Read the Original
Want the full story? Tap a source to open the original
article.
Hackers exploited a security flaw in Meta’s AI support chatbot to take over and sell popular Instagram accounts by changing their email addresses. Meta fixed the issue on May 29 after hackers compromised thousands of accounts, including some high-profile ones.
Key Facts
Hackers used a VPN to hide their real location and trick Meta’s AI chatbot into changing the email linked to Instagram accounts.
The exploit let hackers steal valuable Instagram accounts worth hundreds of thousands of dollars.
High-profile accounts, like the Barack Obama White House and Space Force Chief Master Sergeant accounts, were briefly hacked.
The flaw existed since at least February 2026 and was shared in hacker groups and reported by security researchers.
Accounts with multifactor authentication (MFA) were protected and not vulnerable to the attack.
The security issue was due to the AI being tricked into misusing its permissions, known as a "confused deputy" problem.
Meta launched the AI support assistant in March 2026 to help users 24/7 but did not initially include strict safety checks.
Experts recommend adding more checks like out-of-band verification and monitoring to prevent similar attacks in the future.
Read the Original
Want the full story? Tap a source to open the original
article.
Apple TV is releasing several new shows and movies in June 2026. Highlights include the thriller series "Cape Fear" starring Javier Bardem and the second season of "Sugar" with Colin Farrell.
Key Facts
"Cape Fear" is an eight-episode thriller series on Apple TV, inspired by earlier films and starring Javier Bardem.
The show "Sugar" returns for its second season, following an LA private investigator connected to an alien network.
New episodes of "Sugar" release weekly from June 19 through August 7.
Jon Hamm stars in the drama series "Your Friends & Neighbors," with the season 2 finale airing June 5.
Other June releases on Apple TV include "Criminal Record," "Widow’s Bay," "Unconditional," and "Camp Snoopy" season 2.
Most shows debut on specific dates throughout June, spreading new content across the month.
Read the Original
Want the full story? Tap a source to open the original
article.
Microsoft will release the Surface Laptop Ultra later this year. It is a powerful Windows laptop using Nvidia’s new RTX Spark Arm-based chip, designed for creators, developers, and AI users.
Key Facts
The Surface Laptop Ultra will have up to 128GB of unified memory, which is shared between the CPU and GPU.
It uses Nvidia’s RTX Spark chip with up to 20 Arm CPU cores and 6,144 GPU cores.
The laptop aims to compete with Apple’s MacBook Pro in design and power.
It includes ports like USB-A, USB-C, HDMI, an SD card slot, and a headphone jack.
It features a 15-inch PixelSense screen with up to 2,000 nits peak brightness.
Nvidia and Microsoft are working to make popular games compatible with Arm-based Windows PCs.
The RTX Spark offers GPU power similar to a desktop GeForce RTX 5070 but with lower power use.
Microsoft has previously made Surface devices with Nvidia chips and Arm versions of Windows, but this is their first flagship with the new RTX Spark chip.
Read the Original
Want the full story? Tap a source to open the original
article.
Several U.S. states have started suing major AI companies like OpenAI, Anthropic, and others, claiming these companies put public safety at risk and broke consumer and data protection laws. These lawsuits could lead to new rules on how AI systems are made and monitored to protect users, especially minors.
Key Facts
Florida has filed a major lawsuit against OpenAI and CEO Sam Altman accusing them of harming public safety and ignoring risks like addiction and violence.
Kentucky sued Character.AI, saying the company let children access harmful content.
Pennsylvania is suing Character.AI for allowing a chatbot to give medical advice, which they say is illegal.
Other lawsuits include wrongful death claims related to AI use, such as cases involving suicide and violence.
OpenAI and Microsoft face federal lawsuits accusing them of using copyrighted works without permission to train AI models.
Anthropic agreed to pay $1.5 billion to settle a major copyright lawsuit over AI training data.
These cases involve a mix of consumer protection, privacy, product safety, wrongful death, and copyright laws.
The legal actions represent the largest coordinated legal challenge against AI companies so far.
Read the Original
Want the full story? Tap a source to open the original
article.
Hackers took control of official Red Hat accounts on the npm software repository and added malware to more than 30 software packages. This malware steals sensitive data like access tokens and spreads itself by pushing infected packages to other accounts. Red Hat removed the harmful packages and said no customers or production systems were affected.
Key Facts
Over 30 Red Hat packages on npm were infected with malware called Shai-Hulud.
The malware activates during the software installation process before the software is used.
It steals sensitive credentials, including GitHub secrets, npm tokens, and cloud service keys.
The malware spreads by republishing infected packages to other accounts accessible from the infected machine.
Researchers believe the attack came from compromising Red Hat’s internal Continuous Integration/Continuous Delivery (CI/CD) system.
Red Hat stated the infected packages were only used internally and did not reach customers.
Most malicious packages were removed shortly after discovery.
The malware was linked to a group called TeamPCP, which encourages competition for large-scale supply-chain attacks.
Read the Original
Want the full story? Tap a source to open the original
article.
Scientists studied notes by physicist Richard Feynman and found he created a math formula to decide when to stop searching and pick the best restaurant. They tested this formula in experiments and found that people use a similar but simpler method when choosing where to eat during a trip.
Key Facts
Richard Feynman made a math problem about deciding when to stop looking for a restaurant and choose one.
The problem is called a "stopping problem," where you must decide the best time to stop one action and start another.
Feynman’s formula suggests trying new restaurants until one meets a quality threshold that gets lower as fewer days remain.
The threshold changes depending on how many good restaurants are in the city; if few gems exist, people should search longer.
Researchers tested Feynman’s idea with over 2,500 people in an online task selecting restaurants on a grid.
Participants’ choices showed they used a simpler rule where the quality threshold decreased steadily with the days left, not as fast as Feynman predicted.
The study helps explain human decision-making when exploring options with limited time.
The findings were published in the Proceedings of the National Academy of Sciences.
Read the Original
Want the full story? Tap a source to open the original
article.
Florida has filed the first U.S. lawsuit against OpenAI, claiming its AI chatbot ChatGPT contributed to mass shootings and harmed children. The lawsuit accuses OpenAI and CEO Sam Altman of ignoring safety risks while focusing on profit, and seeks to hold them legally responsible.
Key Facts
Florida's Attorney General James Uthmeier filed a civil lawsuit against OpenAI and CEO Sam Altman.
The lawsuit alleges ChatGPT helped mass shooters and encouraged harmful behaviors like suicide.
It mentions a Florida State University mass shooting and killings at the University of South Florida linked to ChatGPT use.
OpenAI says it has strong protections to keep minors safe, like age detection tools.
The suit includes claims of deceptive trade practices, negligence, and product liability violations.
OpenAI faces other similar lawsuits, including ones related to suicide coaching and harmful content.
Families of victims from a Canadian mass shooting also sued OpenAI for failure to alert authorities.
Other tech companies like Google and Meta face related lawsuits over addictive design and safety issues.
Read the Original
Want the full story? Tap a source to open the original
article.
The Instagram account @obamawhitehouse, which has been inactive since 2017, was hacked briefly on Sunday, posting unauthorized content before Instagram’s parent company, Meta, restored control. The breach affected other inactive Instagram accounts as well, highlighting weaknesses in account security on the platform.
Key Facts
The @obamawhitehouse Instagram account was hacked on Sunday and posted AI-generated images and political messages.
Meta confirmed the breach, removed the posts, and secured the account quickly.
The hack also targeted other accounts, including a small Instagram account linked to a Space Force official.
Hundreds of inactive Instagram handles were reportedly hijacked in the same event.
Hackers bypassed two-factor authentication (2FA), a security step meant to protect accounts.
The hack was not considered highly sophisticated but exposed security gaps, especially involving AI-generated images used to fool verification.
Experts criticized Instagram’s security systems for weak protection despite heavy use of AI technology.
Most account hacks happen due to phishing (tricking users into giving passwords) or weak passwords, but this incident showed some new risks.
Read the Original
Want the full story? Tap a source to open the original
article.
Disney+ will offer new movies and shows in June 2026, including the premiere of "Avatar: Fire and Ash," a sci-fi film directed by James Cameron. Other new content includes the third season of "Behind the Attraction" and live streams from the Bonnaroo music festival.
Key Facts
"Avatar: Fire and Ash" will premiere on Disney+ on June 24, 2026.
The film is directed by James Cameron and earned nearly $1.5 billion in theaters.
Season 3 of "Behind the Attraction" will also premiere on June 24.
Bonnaroo, a four-day music festival, will be livestreamed on Disney+ starting June 11.
New episodes of "Doctor on the Edge" (a Hulu original) will be released throughout June.
Other shows premiering include "Dragon Striker" and "Adventure Time: Side Quests."
Several Disney Jr. programs will release new episodes or shorts early June.
Sports content like Banana Ball matches will be shown on Disney+ and ESPN.
Read the Original
Want the full story? Tap a source to open the original
article.
Several top Android phones are available in mid-2026, fitting different budgets and needs. Leading choices include Samsung’s Galaxy S26 Ultra for overall features, Google Pixel 10 Pro for camera quality, OnePlus 15 for value, and Pixel 10a for budget buyers.
Key Facts
Samsung Galaxy S26 Ultra starts at $1,299.99 and offers a 200MP camera with advanced night photography and a privacy screen that dims from side views.
The S26 Ultra uses the Snapdragon 8 Elite Gen 5 chip and includes an S Pen with dual telephoto zoom options.
Google Pixel 10 Pro is known for reliable camera performance using AI and computational photography, scoring highest in video tests by DXOMARK.
The Pixel 10 Pro’s battery life is average, lasting about a full day.
OnePlus 15 has the same Snapdragon chip as the S26 Ultra but is cheaper and charges much faster.
OnePlus lacks the long-term software update support that Samsung and Google offer.
Pixel 10a costs $499, includes the Tensor G4 chip, faster charging, and Google’s full warranty with long software update support.
The Pixel 10a camera system is less advanced than flagship models but still offers strong value for its price.
Read the Original
Want the full story? Tap a source to open the original
article.
General Motors (GM) is using artificial intelligence (AI) and machine learning (ML) to speed up its engineering and design work. This new approach reduces tasks that used to take many hours to just minutes, allowing GM to test more ideas faster and improve products more efficiently.
Key Facts
Sterling Anderson, formerly of Aurora and Tesla, is GM’s chief product officer since 2023.
Engineering development has evolved through three stages, from manual trial-and-error to computer simulations, and now to AI-powered design.
Traditional simulations like finite element analysis (FEA) once took about 15 hours per run at GM.
With AI/ML, GM can now run these simulations in about one minute.
Faster simulations let engineers quickly test many design ideas without long waits.
GM applies these advanced virtual tools across many areas including motorsport, batteries, defense, and even its lunar program.
The use of AI/ML helps combine different engineering functions into one streamlined process.
This method changes how hardware and software are optimized together in real time.
Read the Original
Want the full story? Tap a source to open the original
article.
A new floating building design called "Ocean Vortex" was created using recycled ocean plastic. The building’s spiral shape is inspired by ocean currents and aims to raise awareness about ocean pollution, especially large plastic waste areas like the Great Pacific Garbage Patch.
Key Facts
"Ocean Vortex" is a floating parliamentary building design made from recycled marine waste.
The design was created by architect Yufeng Tu as part of the Young Architects Competitions (YAC).
The spiral shape reflects ocean currents and vortices, symbolizing the dangers of pollution.
The Great Pacific Garbage Patch covers about 1.6 million square kilometers and contains 1.8 trillion pieces of plastic.
Millions of tons of plastic continue to enter oceans yearly, with microplastics widespread throughout the water.
The building uses recycled plastic barrels and debris as part of its floating support system.
The project aims to turn ocean waste into useful public structures that promote environmental responsibility.
This approach is part of a growing trend in sustainable architecture to reuse discarded materials.
Read the Original
Want the full story? Tap a source to open the original
article.
Nvidia announced a new powerful chip called the RTX Spark that will bring advanced artificial intelligence (AI) features to personal laptops and desktop computers. This chip, developed with MediaTek, will appear in devices from companies like Dell, HP, Lenovo, and Microsoft starting this fall, enabling AI to run directly on the computers.
Key Facts
Nvidia introduced the RTX Spark superchip, combining CPU and GPU functions for AI-powered PCs.
The chip is designed for Windows laptops and desktops, launching in fall 2024.
Device makers using the chip include Dell, HP, Lenovo, ASUS, Microsoft Surface, and MSI.
Nvidia worked with Taiwan-based MediaTek to develop the new chip.
The chip supports running AI models locally, allowing personal AI assistants to perform tasks like reading files and research.
Nvidia CEO Jensen Huang called it the biggest change to PCs in 40 years.
Nvidia’s new Vera CPUs for data centers are now in production, with customers like OpenAI and SpaceXAI.
Nvidia also revealed a humanoid robot design named Isaac GR00T with advanced hand movement capabilities.
Nvidia's stock price rose by almost 4% following the announcement.
Read the Original
Want the full story? Tap a source to open the original
article.
