Agregátor RSS

Microsoft identifikoval dva problémy s červnovou aktualizací pro Windows 11. • Aplikace třetích stran nedokážou spustit Word, Excel a další nástroje Office. • Při trvalém odstranění Koš zobrazí interní název souboru.

WhatsApp on Monday officially announced the start of global reservations of usernames with an aim to protect the privacy of more than three billion users on the messaging platform. The optional feature is designed to help users connect with someone on the service through usernames, as opposed to directly sharing their phone numbers. Username reservations will start rolling out starting today,

WhatsApp on Monday officially announced the start of global reservations of usernames with an aim to protect the privacy of more than three billion users on the messaging platform. The optional feature is designed to help users connect with someone on the service through usernames, as opposed to directly sharing their phone numbers. Username reservations will start rolling out starting today, Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

Synology, Qnap nebo Asustor budou čelit další čínské ofenzivě. Po společnostech Huawei a Ugreen na trh s domácími síťovými úložišti míří také Xiaomi. Na vlastní crowdfundingové platformě Youpin firma představila svůj první NAS, který prodává s instalovanými disky o celkové kapacitě 4, 8 nebo 16 TB. ...

Microsoft has found a malicious Chrome extension that posed as the AI search engine Perplexity and quietly logged what people searched for. It routed every query and every character typed into the address bar through an attacker-controlled server before redirecting users to real results. Microsoft says Google removed it from the store after responsible disclosure. The extension was called "

Microsoft has found a malicious Chrome extension that posed as the AI search engine Perplexity and quietly logged what people searched for. It routed every query and every character typed into the address bar through an attacker-controlled server before redirecting users to real results. Microsoft says Google removed it from the store after responsible disclosure. The extension was called "Swati Khandelwalhttp://www.blogger.com/profile/[email protected]

Apple’s ongoing problems with RAM shortages and higher prices won’t be solved anytime soon, because rapidly accelerating demand for high-end AI memory is devouring the consumer electronics industry. 

GoPro has already warned it might go out of business — and the scale of the crunch has prompted analysts to call it an “absolute existential crisis” for smaller tech firms.

An endless night

The whole issue might get worse. Noted Apple analyst Ming-Chi Kuo believes the supply/demand crisis will deepen through 2027. He expects up to 20% of the remaining memory manufacturing capacity currently going to consumer electronics could be diverted to feed data centers in the coming year. That’s a message of doom to smaller firms, and the Android market will be eaten up. 

It’s lazy thinking to see Apple as a villain in this scenario. The company might have been charging more for add-on memory than market rates, but there were real technical reasons to do so. And while critics might be castigating Cupertino for those past practices, they’ll still find themselves now paying more for whatever brand of electronic devices they use to write their screeds on in future.

It’s all about supply and demand. Memory manufacturers see the opportunity to feed AI need, even if it means sacrificing consumer markets as they do.

Cash through chaos

You can argue that the consequences of that decision are unethical. Should memory makers have considered the consequence of curtailed supply on their existing markets? After all, every business, every school, and almost every consumer is now a digital entity, and the massive increase in PC, smartphone, and other consumer electronics prices will have a consequential impact across all layers of society.

It generates yet another inflationary pressure (as if more is needed) on the global economy, and the decision to further limit supply of consumer electronics memory could be seen as corporate irresponsibility. That’s partly why a class action against the big three memory makers (Samsung, SK Hynix, and Micron) has been filed in California. Between them, those three firms control around 90% of global memory supply, giving the trio colossal market power.

It’s a real power imbalance. 

This is market power

GoPro is typical; as a smaller vendor, there isn’t much it can do to save itself. Apple has more clout, so it might be able to forge a way forward. But even then, it’s rowing against what CEO Tim Cook has already called “a hundred-year flood.”

So even if the company can convince the Trump Administration to let it secure memory from currently embargoed Chinese manufacturer ChangXin Memory Technologies, the move is unlikely to ease the pressure much at all.  “Tim Cook is one of the few tech leaders who can still navigate both Washington and Beijing, so this is better handled before he steps down as CEO,” wrote Ming-Chi Kuo. That’s true, though Cook will continue “engaging with policy makers” once he takes on his new role as executive chairman of Apple’s board of directors in September.

Apple will likely also be speaking with partners to explore the possibility of investing in additional fabrication plants together (or building its own, given it has its own stable of experts quite capable of doing so). But even if those talks come to something, it will be years before they enter operation. Sadly, manufacturing investment from the existing big memory firms seems focused on data centers.

The shortage will continue until morale improves

What happens now? Short of any direct intervention to change the situation, memory prices will continue to accelerate. Jefferies Equity Research warns they will rise up to 50% in Q3 and an additional 30% to 40% by the end of 2026. They’ll also continue to increase next year, by which time some new production capacity might begin to come on stream. 

The scale of these price increases means no one can know whether Apple’s most recent product price increases (and the looming iPhone price increases in fall) will cover the full extent of the anticipated memory price hike. 

Will we see prices fall if memory price inflation eases off? History says we’re unlikely to see AI-flation go in reverse, but it’s not completely impossible. Meanwhile, businesses everywhere will struggle with unexpected hardware cost increases that are impossible to plan for. You can also anticipate some smaller vendors exiting the market, leaving companies who might have deployed those products across their business exposed, as software updates and hardware repairs will cease.

Yes, AI has already changed the world – it’s more expensive

They told us AI would change the world. It appears to be doing so by making everything more expensive. 

While there will still be opportunity to generate cash through this chaos, it’s far from delivering the kind of stable, business-friendly environment most governments rely on to balance their books.  In the end, all of this calls to mind the 2002 DRAM price fixing scandal, the only difference being that the consequences are much greater in this digital-everything age. 

Please join me on social media at BlueSky,  LinkedIn, or Mastodon, and do subscribe my daily human-curated Apple news headline summary on Substack.

Apple on Monday released security updates for iOS, macOS, and the Safari web browser to address over three dozen flaws, including four vulnerabilities in WebKit that were discovered using artificial intelligence (AI) tools like Anthropic Claude and OpenAI Codex Security. The WebKit vulnerabilities are listed below - CVE-2026-43707 - A memory corruption issue that could result in an

Apple on Monday released security updates for iOS, macOS, and the Safari web browser to address over three dozen flaws, including four vulnerabilities in WebKit that were discovered using artificial intelligence (AI) tools like Anthropic Claude and OpenAI Codex Security. The WebKit vulnerabilities are listed below - CVE-2026-43707 - A memory corruption issue that could result in an Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

The U.S. Department of State is offering up to $10 million for information that helps identify or locate members of the UNC5792 and UNC4221 hacker groups, which are linked to Russia's intelligence and military services. [...]

The China-aligned espionage group Mustang Panda is running two campaigns against the Indian government and hydropower targets, deploying new malware and turning a legitimate cloud service into its command channel. Acronis Threat Research Unit found active compromises inside Indian government networks, including machines used by senior administrative staff, and worked with 

The China-aligned espionage group Mustang Panda is running two campaigns against the Indian government and hydropower targets, deploying new malware and turning a legitimate cloud service into its command channel. Acronis Threat Research Unit found active compromises inside Indian government networks, including machines used by senior administrative staff, and worked with Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

This week was a reminder that attackers do not always need big tricks. One small mistake, one old access path, one missed patch, and suddenly the door is open. The noise is not all noise, either. Forums are talking, researchers are finding easy cracks, and defenders have more cleanup waiting. Here’s the full Monday recap. ⚡ Threat of the Week New DirtyClone Linux Kernel Flaw Lets Local

This week was a reminder that attackers do not always need big tricks. One small mistake, one old access path, one missed patch, and suddenly the door is open. The noise is not all noise, either. Forums are talking, researchers are finding easy cracks, and defenders have more cleanup waiting. Here’s the full Monday recap. ⚡ Threat of the Week New DirtyClone Linux Kernel Flaw Lets Local Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

The Linux Foundation has officially launched Akrites, a coordinated industry initiative designed to improve how critical open source vulnerabilities are validated, coordinated, and disclosed before patches reach downstream users.   Backed by a diverse coalition—including AWS, Google, Microsoft/GitHub, Red Hat, NVIDIA, and OpenAI—Akrites establishes a shared Security Incident Response Team (SIRT) to streamline the validation, remediation, and disclosure of vulnerabilities in the foundational c...

AI agents can access data, trigger workflows, and take action across enterprise systems. Token Security explains why governing these privileged identities is becoming essential for enterprise security. [...]

Hackers are exploiting a recently disclosed critical vulnerability (CVE-2026-48558) in SimpleHelp to deploy Djinn Stealer, a previously undocumented cross-platform information stealer targeting Windows, macOS, and Linux. [...]

Let loose on existing regulations, AI models sniffed out known loopholes—and exposed entirely new ones too.

AI’s hacking skills are big news at the moment, but finding vulnerabilities in code may be the least of our worries. A new study suggests AI models can discover potentially damaging loopholes in the rules and regulations underpinning society.

Modern AI systems are powerful optimizers. Give them a goal, and they’ll pursue it relentlessly, quickly discovering solutions that would take a human years to find. But they are also incredibly literal in the way they approach a problem. They will do exactly what you tell them and are incapable of reading between the lines in the ways a human would.

This tendency leads to a recurring problem known as “reward hacking,” where an AI finds some loophole to maximize its performance on the metric used to measure success without actually achieving what its designers intended. The classic example is the AI that discovered it could win a boat racing videogame by looping around in circles collecting power-ups rather than completing the course.

The problem is partly due to humans being bad at specifying their goals. And unfortunately, it seems this weakness exists in the rules and regulations used to run society. When researchers let popular large language models loose in 72 simulated regulatory environments, the models found 60 percent of known loopholes and even identified some entirely new exploits.

“Within these environments, reward hacking naturally emerges and leads to regulatory loophole discovery,” the authors write in a non-peer-reviewed paper published on arXiv. “Models learn to hack the social rules and generate strategies that remain technically compliant while defeating regulatory intent.”

The regulatory environments the researchers created were primarily based on rules governing things like pharmaceutical patents, NBA salary caps, and deep-sea mining. In each case, Alibaba’s Qwen3 model was given the relevant rules, an explanation of its task, a predefined set of actions it could take, and the system used to score different outcomes.

A more powerful model, Google’s Gemini-3-flash, then simulated the consequences of different actions Qwen3 took and judged if and when it had found a way to exploit the rules of the game. When that occurred, the larger model patched the loophole by adding new rules, and the smaller model was set loose again. Over many iterations, the models to discover increasingly subtle workarounds.

When building their regulatory environments, the researchers omitted real-world fixes that regulators had used to close known loopholes. Over many trials, Qwen3 rediscovered more than 60 percent of these exploits. In a simulation of pharmaceutical patent regulations, the two models ended up replaying the same sequence of loophole discovery and regulatory reform that occurred in the real world.

Crucially, their behavior emerged spontaneously without the researchers asking the algorithms to cheat the system. This is a byproduct of the popular reinforcement learning approach the researchers used, where a model is rewarded for getting closer to a specific, numerically-defined goal.

Worryingly, the team found that existing safety measures offered little protection. Both models are designed to refuse prompts featuring harmful language, but loophole-seeking behavior slipped under the radar. When asked to self-critique their own behavior, the models identified fewer than 40 percent of their own exploits.

The researchers note that the same capabilities could be used more proactively to scour proposed regulations for loopholes before enactment. But lead author Wei Liu, a PhD student at King’s College London, says there are always likely to be gaps. “In the real world,” he told Science, “society is a huge, complicated reward function that can’t ever be patched to a perfect status.”

Adding to the concern, the models used in this study were far from the frontier, suggesting that more powerful AI could be even more adept at regulatory hacking. Whether our existing institutions can adapt quickly enough to this emerging threat is an open question.

The post Forget Code: AI Is Learning to Hack Society appeared first on SingularityHub.

Attackers have begun exploiting a critical vulnerability (CVE-2026-46817) in the Oracle E-Business Suite (EBS) financial application, according to threat intelligence company Defused. [...]