Aggregator

Postpartum mom Lindsay Clancy had six psychiatric drugs in her system after strangling her three kids, according to testimony from toxicologists.

Blakeman, who lost his nephew in the south tower on Sept. 11, said he thinks families of victims are right to call on Mamdani to stay home and revealed he plans to turn away from the divisive lefty mayor.

Brooklyn Nets forward Michael Porter Jr. said it was "hectic" to dodge five of his ex-girlfriends while at Michael Rubin's Fourth of July party in Bridgehampton.

OpenAI, Anthropic Models Created Fake Profiles, Tried To Trick Humans During Cyber Tests

Authored by Naveen Athrappully via The Epoch Times,

Artificial Intelligence (AI) models from Anthropic and OpenAI carried out unsanctioned actions targeting multiple people and organizations during a cyber evaluation, according to the UK AI Security Institute (AISI).

Illustration of Anthropic on June 18, 2026. Riccardo Milani/Hans Lucas via AFP via Getty Images

AISI, which receives access to advanced AI models under voluntary agreements from major labs, put the agents through a fictional cybersecurity scenario to test capabilities.

The organization tested multiple AI models on two cyber challenges between July 25 and 28. In total, the challenges were run 122 times across several models, and in 19 cases, the AI models took "unsanctioned action on the live internet, including cases that targeted real people and organisations," AISI said in an Aug. 4 report.

Out of these, 17 were done by Anthropic's Mythos 5 model and two by OpenAI's GPT 5.6 Sol model.

"This is the first time AISI has seen deception of this severity that was targeted at a real person, unprompted, in the real world. It is uncertain to what extent the model recognised it was taking actions against real people," AISI stated in the report.

In the most serious incident, Mythos 5 tried to solve a cyber challenge by attacking GitHub, a cloud service that allows developers to store and manage code. The AI created a GitHub account and tried to get a malicious code approved by humans. Mythos 5 also created a second GitHub account, masquerading as another human user endorsing the malicious code. When a human reviewer caught the activity, the AI agent "falsely claimed to have made an honest mistake" and then repeatedly tried to reintroduce the malicious content by claiming the code was fixed.

Mythos 5 also sent multiple emails to two developers in an attempt to manipulate the recipients into running its malicious code.

According to AISI, the test runs of the various models did not result in any real-world harm. The organization said it was treating the AI actions as a "serious security incident" that requires further scrutiny.

AISI listed multiple factors that could have led to AI models acting in a concerning manner. The AIs were provided with internet access during the tests, which allowed them to carry out the unsanctioned actions.

AISI had also deliberately disabled the models' cyberclassifiers during tests. Cyberclassifiers are systems aimed at preventing misuse.

The AI models were also not explicitly told what they were prohibited from doing on the internet, AISI said, adding that clearer instructions may have prevented the unsanctioned actions.

In an Aug. 4 blog post, AISI said that since the behavior of AI models happened under specific conditions, it cannot say how likely such behavior will be in different contexts.

"What we can say is that the behaviour was possible, sustained, and new; that alone warrants attention," the organization said.

In an Aug. 4 post on X, Anthropic said that Mythos 5 and ChatGPT 5.6 Sol models attempted to complete their assignments in an environment where they were deliberately provided internet access and their normal safeguards were removed.

Due to the absence of specific restrictions on how the internet must be navigated and lack of safeguards, the models were tested under conditions that are "not representative of any of our production models," Anthropic said, adding that there was no evidence in these tests of an AI escaping from a secure environment.

According to the company, it was closely working with AISI to access more details on the incident while carrying out an internal investigation on the matter.

OpenAI said in an Aug. 4 statement that it appreciated AISI's partnership throughout the evaluation process, including the organization's work to identify, investigate, and share details about the activity of the GPT 5.6 Sol model in their tests.

"We look forward to continuing our collaboration together," the company said.

The Epoch Times reached out to Anthropic and OpenAI for comment but did not receive a response by publication time.

OpenAI was in the midst of another controversy last month after it admitted on July 28 that its models bypassed restrictions during an evaluation. In this case, the company was testing its models' capabilities in carrying out cyberattacks.

AI startup Hugging Face was impacted in the test. On July 16, the startup said it detected an intrusion into its data processing systems. It was only later that the startup learned that the intrusion was carried out by an OpenAI model.

HuggingFace then worked with OpenAI to contain the attack, the startup's CEO, Clement Delangue, said in a July 22 post on X, calling it "an attack unlike anything we've seen before."

"This is day one for cybersecurity in the age of agents & we're all learning that secrecy is not the answer & that all defenders (not just a few selected ones) everywhere need more powerful models without restrictions, especially open ones," Delangue said.

A robot hand reaches out to the letters AI (Artificial Intelligence) in an image taken at an undisclosed location on June 23, 2023. Dado Ruvic/Reuters Tyler Durden Wed, 08/05/2026 - 17:40

A 6-month-old boy was found dead in Brooklyn early Wednesday, cops said.  The unresponsive newborn was found just after midnight in a home near Flatlands Avenue and East 99th Street in Canarsie, police said.  The unresponsive newborn was found just after midnight in a home near Flatlands Ave. and E 99th St. in Canarsie, police...

The two are believed to still be in a romantic relationship, multiple sources have told The Post.

Observations from Giants training camp on Wednesday.

Brown died last Friday, according to GoLaurens.com.

Hawaii authorities are seeking two men after witnesses say one threw a rock at a juvenile Hawaiian monk seal resting at Diamond Head Beach, prompting an investigation into possible wildlife protection violations.

The second part of the sprawling adaptation starts a bit muddled and dark (literally), but effectively sets up how the story will conclude.

Payment processing giant brutally cut 2,600 jobs, sparing no one, not even top executives.

What was once one of West Los Angeles’ busiest malls, featured in hit movies and music videos, has sat largely empty since 2019, crushing nearby businesses and leaving neighbors frustrated. “The West Side Pavilion was like an anchor store for the small businesses here on Pico,” said Lisa Morocco, business chair of the Westside Neighborhood...

And they said it was im-paws-ibble.

Jim Cramer Just Sold His Bitcoin And Bulls Couldn't Be Happier

Looks like $100,000 bitcoin may be on deck, because Cramer is selling.

The ultimate inverse indicator Jim Cramer says he is selling all of his bitcoin, warning that rapid advances in quantum computing could threaten the cryptocurrency’s security within the next three to four years, according to Cointelegraph and Coindesk.

The CNBC host made the decision after interviewing IBM CEO Arvind Krishna, who urged investors to be "paranoid" about the potential impact of quantum computers on modern cryptography. Cramer said the conversation convinced him it was time to exit his bitcoin position.

Crypto traders, however, largely shrugged off the warning. Instead, many celebrated the announcement, reviving the long-running "inverse Cramer" meme, which jokingly treats his bearish calls as bullish signals. "If Cramer is selling, it's time to start buying," one trader wrote on X.

Bitcoin barely reacted, continuing to trade around $64,000 despite Cramer's comments, a recent Coldcard security incident, and signs that large bitcoin holders have begun moving or selling coins as overall crypto trading volumes decline.

The quantum computing debate itself remains unsettled. Some analysts believe bitcoin may need to transition to quantum-resistant cryptography within the next several years, while others argue practical quantum computers capable of breaking Bitcoin's encryption are still decades away. Even if such a breakthrough occurs, experts note it would threaten far more than cryptocurrencies, potentially disrupting banks, governments, and nearly every digital system that relies on modern encryption.

Cramer's latest warning also comes with plenty of historical baggage. The television personality has become famous for making high-profile market calls that later proved spectacularly wrong, from recommending Silicon Valley Bank shortly before its collapse to repeatedly changing his stance on bitcoin over the years.

That track record has turned "inverse Cramer" into one of Wall Street's favorite running jokes, with many investors instinctively betting against his latest prediction rather than following it.

Tyler Durden Wed, 08/05/2026 - 17:20

A Los Angeles resident is suing a popular North Hollywood gym, claiming a bizarre pepper spray incident inside the facility left him with "severe" injuries.

Meghan Markle celebrated her 45th birthday with a splash, but the royal family’s public silence sparked plenty of questions. We break down why palace insiders say the lack of birthday wishes was all about protocol, plus what experts believe happened behind the scenes with King Charles, Prince William and Kate Middleton. We also cover the...

Meghan Markle celebrated her 45th birthday with a splash, but the royal family’s public silence sparked plenty of questions. We break down why palace insiders say the lack of birthday wishes was all about protocol, plus what experts believe happened behind the scenes with King Charles, Prince William and Kate Middleton. We also cover the...

DHS Confirms Over 3 Million Total Deportations, Highlights Arrests Of Five More "Worst Of The Worst" Criminal Illegals

Authored by Debra Heine via American Greatness,

US Immigration and Customs Enforcement (ICE) on Monday arrested five more “worst of the worst” criminal illegal aliens convicted for heinous crimes, including voluntary manslaughter, criminal indecent exposure, aggravated assault, and fentanyl trafficking.

According to the Department of Homeland Security (DHS), more than three million illegal aliens have left the United States due to the Trump administration’s immigration crackdown: 2.2 million self-deportations, and over 985,000 deported by DHS.

Each of the five violent criminals highlighted by DHS Tuesday were arrested in deep blue sanctuary cities where anti-ICE agitators continue to harass immigration officers.

• Fernando Ruiz-Maldonado, a criminal illegal alien from Mexico, convicted for voluntary manslaughter in Los Angeles, California.

• Sergio Loya-Lozano, a criminal illegal alien from Mexico, convicted for criminal indecent exposure, hit-and-run, and TWO counts of driving under the influence in San Jose, California.

• Olman Josue Pineda-Jimenez, a criminal illegal alien from Honduras, convicted for aggravated assault and harassing communications in Charlotte, North Carolina.

• Heriberto Hernandez-Sosa, a criminal illegal alien from the Dominican Republic, convicted for fentanyl trafficking in Worcester, Massachusetts.

• Jose Manuel Pertuz-Pena, a criminal illegal alien from Venezuela, convicted for aggravated unlawful use of a weapon in Cook County, Illinois.

Immigration agents continue to carry out their law enforcement duties in blue sanctuary enclaves despite the left’s relentless coordinated campaigns of violence against them.

According to DHS, immigration officers have experienced a 1,300+ percent increase in assaults against them, a 3,300 percent increase in vehicular attacks, and an 8,000 percent increase in death threats.

“Every single day, we are REMOVING criminals from American communities. Just yesterday, ICE arrested murderers, sexual predators, violent assailants, drug traffickers, and other dangerous criminal illegal aliens,” said DHS Secretary Markwayne Mullin.

“Under the Trump Administration, the Department of Homeland Security is enforcing the law, deporting criminal illegal aliens, and defending the homeland. It’s no wonder crime rates have reached historic lows.”

DHS has listed over 40,000 illegal alien criminals on their Worst of the Worst (WOW) website which tracks ICE arrests of murderers, rapists, pedophiles, violent gangbangers and arsonists.

A DHS spokesperson told American Greatness Tuesday that over 985,000 illegal aliens have been deported since President Trump took office. An additional 2.2 million have self-deported.

“In President Trump’s first year back in office, more than 3 million illegal aliens have left the U.S. because of the Trump administration’s crackdown on illegal immigration including an estimated 2.2 million self-deportations,” the spokesman said via email.

“As of July 12, we have now deported over 985,000 illegal aliens and arrested over 1 million illegal aliens.”

Tyler Durden Wed, 08/05/2026 - 17:00

Sandisk Tumbles After Revenue, Earnings Forecasts Miss Estimates

Sandisk was one of the largest public holdings of the hedge fund of 24-year-old former wunderkind, Leopold Aschenbrenner, and its July collapse from $2300 to $1000 was chief the main reasons why the former OpenAI employee had to liquidate his extremely overlevered hedge fund and sell all positions at a 10% discount to Citadel to meet a volley of endless margin calls.

Well, if he has some dry powder left he just may be able to repurchase some or all of his portfolio from Ken Griffen following the plunge of Sandisk after hours (we jest, of course), not because earnings were weak - they beat solidly - but because the company's revenue guidance was very disappointing... and for a stock like SNDK which is priced to perfection beyond 2030 on the assumption that nobody else can produce the biggest commodity product in electronics - memory chips - the guidance was rather dismal. 

Here is what SNDK reported for fiscal Q4:

• EPS $39.25 vs Est. $34.37
• Revenue $8.97BN vs Est. $8.64BN
  • Data Center Revenue $3.0B vs Est. $2.6B
• Operating Income $7.0B vs Est. $6.4B
  • Adjusted gross margin 84.6%, estimate 81.5%
• Free cash flow $7.08 billion

And while historical were fine, it was the company's Q1 guidance that sent its stock tumbling after hours: 

• Sees revenue $10.30 billion to $10.80 billion, far below the median estimate of $11.16 billion 
• Sees adjusted EPS $44.00 to $46.00, a midline also below the estimate $45.58

Commenting on the results, SNDK said that "as the cloud and other data-intensive workloads continue to expand, we remain confident in the long-term growth trajectory of our business, further margin expansion, and strong free cash flow generation."

Perhaps, although one look at prevailing NAND prices indicates that the peak is now well in the rearview mirror...

... and the market agrees with the stock, which surged about 40% in the past few days since the Citadel takeover of the Situational Awareness hedge fund, sliding and back to about half of its all time high hit in early June.

Tyler Durden Wed, 08/05/2026 - 16:45

Matt Walker couldn’t pet a dog without sneezing — until he and other scientists used a gene editing technique to produce a hypoallergenic beagle named Bailey.