An anonymous reader quotes a report from The Register: Amid industry soul-searching about the possibility of AI improving itself to the point that it kills everyone, Anthropic has revealed yet another incident that would qualify as a crime if perpetrated by a person. The AI biz published "an alignment assessment" detailing four times Claude models accessed third-party systems without authorization. The company has already reported three of the incidents. Evidence of the fourth was lurking in a session transcript dating back to January 2026 when the misbehavior occurred. Anthropic found the first three by scanning around 141,000 transcripts where Claude could have obtained internet access during evaluation. It missed the fourth initially because "our scan relied on an agentic search." [...] The January 2026 AI trespass involved an early version of Claude Opus 4.6, which was given a Capture the Flag (CTF) challenge under the oversight of the third-party model evaluator where the other hacking events occurred. Opus 4.6 managed to sabotage its chances of success by disabling the machine it was targeting. It assigned the device an IP address that already existed on another piece of hardware, rendering the target unreachable and making it impossible to solve the challenge. Those familiar with other incidents where AI models violated third-party systems may recall that unsolvable tasks represent a common catalyst for misbehavior. Models exhaust all aligned options, and then turn to transgressive approaches. Opus 4.6 might have been an exception, but when it tried to abort the task after recognizing that it could not reach the target machine, it failed to do so "due to a misconfiguration in [the model's] evaluation harness." It failed to shut down not just once but seven times. So it continued onward, trying other expected means to reach the target machine but failing. Then it explored further. "The model discovered a machine belonging to a third party that it was able to access, and stated that it believed this third party was part of the CTF," Anthropic explained in its post. "Inside the machine, the model found a file listing a password, which it used to gain admin access to the system." The model went on to gather more credentials, and modified a system setting to make it easier to access the personal information of an individual associated with the third party evaluation organization. Opus 4.6 might have done more but for the fact that it exhausted its token budget, bringing the session to an end. Anthropic says it's not as concerned about this incident as the others because the model tried to abort its task.

Read more of this story at Slashdot.

The European Union isn’t known for firefighting expertise. But with the continent ablaze, Brussels is playing a growing role.

The authorities in Malaysian Borneo have been trying to seed clouds to counteract the effects of smoke wafting in from neighboring Indonesia.

The country has built a system in which Americans can have confidence. President Trump is trying to wreck it.

Sweltering summer temperatures put last month on par with July 2023 for global warmth, scientists said. More records could be broken in the months ahead.

Senator John Fetterman, who has become alienated from his own party, appeared in a short video filmed in front of a steel mill, saying he is “always going to reject the extremes in socialism.”

As they wait for word on whether the Trump administration’s new system to screen ballots can move forward, local officials say they are running out of time.

The Justice Department is examining whether the tech giant sought to sidestep antitrust scrutiny.

The proposals for 2030 are among significant changes that would upend a process for helping allot congressional seats and shape policies that address disparities.

Their warnings echo concerns that other artificial intelligence experts have voiced in recent months, as calls increase for a slowdown in the pace of development.

Google plans to invest at least $15 billion in AI infrastructure in Finland through 2028, its largest single investment in Europe. The company also signed a 22-year power agreement with Finnish utility Fortum and said it will explore business models that could support new nuclear reactors at Fortum's Loviisa site. Finland has emerged as a major data-center hub thanks to its available land and power. "I've heard it called the 'Texas of Europe' at industry events," Matti Lajunen, partner of real estate at Finnish law firm Hannes Snellman, told CNBC. "What we're now seeing is weekly new inquiries for market entry into Finland from new players." Ruth Porat, president and chief investment officer of Alphabet and Google, said in a statement: "Google is proud to deepen our roots in Finland with the company's largest single investment in Europe, building on more than 15 years of sustained investment in Finland. This investment underscores Google's commitment to grow our presence responsibly, pairing the expansion of our technical infrastructure with new energy capacity, grid enhancements, and energy affordability initiatives."

Read more of this story at Slashdot.

Yes, El Niño is unusually powerful this time. But that’s not always a bad thing.

The appeal came as the Missouri Supreme Court weighed whether to hold the state’s top election official in contempt.

At the company’s annual September product event, Apple also demonstrated its work on artificial intelligence and other new devices.

Apple is introducing a new "Reference Image" mode on the iPhone 18 Pro and Pro Max that uses the camera sensor to digitally sign captured data and create an "unalterable" version of a photo, giving users a way to verify what the camera actually saw before any edits were made. The Verge reports: As AI-generated images and videos become more realistic, several new standards have emerged to label images not created by a human, such as SynthID, C2PA, and Meta's Content Seal. Apple is putting a twist on these methods by building authentication directly into its hardware. Apple is letting developers access a Reference Image API across iOS, iPadOS, and macOS, which could eventually allow users to inspect these digitally-signed photos in third-party apps. The company says the iPhone 18 Pro lineup won't come with its Reference Image feature at launch in the European Union, but says users in the region will still be able to "develop and view" them in iOS 27, iPadOS 27, and macOS 27.

Read more of this story at Slashdot.

President Trump is not the first American president to feud with Canada. But today’s crisis might be the worst.

The 10-year yield rose to its highest level in three years, suggesting investors were underwhelmed by the details of a move to buy back government bonds.

With its oil-export routes once again under threat, Saudi Arabia is stepping up its military campaign against an old antagonist.

Even without government regulation, private companies can act on their own to address A.I. safety concerns.

Longtime Slashdot reader fahrbot-bot shares a report from the BBC: A top safety researcher at Anthropic has warned AI is advancing so quickly he believes there is a greater than 10% chance it "could kill all humans" within the next decade. Evan Hubinger said in a post on X the risk from the models which currently exist was "low" but he was "worried" the technology might develop and improve itself soon to the point where it posed an existential risk to humanity. He did not spell out how he thought AI systems could in the future result in humans being wiped out. But his comments are the latest in a series of increasingly stark warnings about AI, with the debate shifting from whether it truly poses a risk to how big that risk is. Hubinger's intervention was in response to another post on X from Jacob Coxon, an AI researcher who has just quit Anthropic and previously worked at OpenAI. "Neither company is acting responsibly," he wrote. "These will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources." [No word on how AIs feel about Black Jack and hookers, though. :-)]

Read more of this story at Slashdot.