9820
Total Links
5721
News
3033
Social
66
Groups
Found 47 links from Ars Technica
Researchers used Claude to hack OpenAI
Researchers from Hacktron AI breached an OpenAI employee’s ChatGPT account by exploiting vulnerabilities in OpenAI’s community forum, hosted by third-party platform Discourse. The attackers used the f...
OpenAI · Anthropic
2
Lawsuit: Amazon refused to give pregnant workers bathroom breaks and chairs
A class-action lawsuit filed in US District Court for the Eastern District of New York alleges that Amazon discriminated against pregnant warehouse employees by denying basic accommodations such as ba...
Willamina Barclay · Kristina Green · Amazon · US District Court for the Eastern District of New York
1
Trump may be forced to reveal secret rules feds use for AI safety testing
Protect Democracy, a nonpartisan nonprofit, has sued four federal agencies to force disclosure of the Trump administration’s secret AI safety review framework. The group says the public and Congress h...
Trump · Greg Casar · Protect Democracy · OpenAI
2
BGP hijack infecting networks caused by a comedy of errors that’s not funny at all
Hackers carried out a supply chain attack by hijacking Internet routing to deliver malware disguised as software updates. The target was Softaculous, a UAE-based company that develops Virtualizor, a p...
Ben Cartwright-Cox · Doug Madory · Softaculous · Hetzner Online
2
Cloudflare turns AI against itself with endless maze of irrelevant facts
Cloudflare announced “AI Labyrinth,” a new anti-scraping feature designed to frustrate unauthorized AI crawlers by feeding them fake, AI-generated pages instead of simply blocking them. The company sa...
Cloudflare · Ars Technica
2
Terabytes of credentials leaked in massive supply-chain attack
A major supply-chain attack on the open source AI development tool LiteLLM exposed terabytes of sensitive credentials belonging to thousands of organizations, including Microsoft, Amazon, Cisco, Samsu...
Kevin Beaumont · Alon Gal · LiteLLM · Microsoft
6
The United States is about to wake up to the threat from China's space program
China’s upcoming Chang’e 7 mission could become a major milestone in the US-China space race. Launching as soon as Sunday aboard a Long March 5 rocket, the mission includes an orbiter, lander, rover, ...
Jared Isaacman · Eric Berger · NASA · United States Congress
3
Now, defenders are embracing the prompt injection, too
Researchers at Tracebit say defenders can turn prompt injection, a tactic usually used by attackers against AI systems, into a security tool. Their new method, called “context bombing,” places malicio...
Andy Smith · Earlence Fernandes · Tracebit · Amazon Web Services
2
Now, defenders are embracing the prompt injection, too
Researchers at Tracebit say defenders can turn prompt injection, a tactic usually used by attackers against AI systems, into a security tool. Their new method, called “context bombing,” places malicio...
Andy Smith · Earlence Fernandes · Tracebit · Amazon Web Services
1
Commodore’s newest gadget is a flip phone that blocks social media and browsers
Commodore, revived last year after Christian “Peri Fractic” Simpson acquired the brand and its trademarks, is expanding beyond retro PCs with a new flip phone called the Callback 8020. The device cont...
Christian Simpson · Peri Fractic · Commodore · Commodore Corporation
1
OpenAI's first branded hardware is... a light-up keyboard?
OpenAI has launched its first branded hardware product, the $230 Codex Micro, amid ongoing speculation about more ambitious devices, including a personalized smart speaker. The Codex Micro is a limite...
Jony Ive · Kyle Orland · OpenAI · Work Louder
2
How to burst the AI bubble: Strike at its roots - Ars Technica
Cory Doctorow’s new book, *The Reverse Centaur’s Guide to Life After AI*, argues that the current AI boom is driven by hype, financial incentives, and a desire to replace human labor rather than by cl...
Cory Doctorow · Ars Technica · Macmillan Publishing
2
Critical Copilot vulnerability allowed hackers to seal 2FA code from users - Ars Technica
Microsoft recently patched a critical vulnerability in its M365 Copilot platform that could have let attackers steal two-factor authentication codes and other sensitive information from users’ emails....
Dan Goodin · Microsoft · Ars Technica
1
Man sues Florida cops over arrest spurred by "93% match" in facial recognition - Ars Technica
A Florida man, Robert Dillon, is suing Jacksonville Beach police and other law enforcement officials, alleging he was wrongly arrested and prosecuted after officers relied on a flawed facial recogniti...
Robert Dillon · Scott O’Connell · City of Jacksonville Beach · Jacksonville Beach Police
2
Claude’s new AI file-creation feature ships with security risks built in - Ars Technica
Anthropic has launched a new “Upgraded file-creation and analysis” feature for Claude that lets users generate Excel spreadsheets, PowerPoint presentations, and other documents directly in chat and th...
Simon Willison · Anthropic · Ars Technica
2
Nobody needs AI to search the Internet, court says in ruling against Google - Ars Technica
A German court has issued a preliminary ruling that could have major implications for AI search tools, finding Google liable for false statements made in its AI Overviews. The case involved two publis...
Google · Ars Technica
When it comes to total water use, AI data centers are a drop in the bucket - Ars Technica
A new Ars Technica report argues that AI data centers’ total water use is small compared with overall water consumption, even though individual facilities can heavily affect local supplies. Amazon sai...
Kyle Orland · Amazon · Google
AT&T and Verizon lose Supreme Court case over fines for selling location data - Ars Technica
The Supreme Court ruled 8-1 that AT&T and Verizon cannot overturn FCC fines for selling customers’ real-time location data without consent on Seventh Amendment grounds. The case centered on $104 milli...
Clarence Thomas · Roberts · AT&T · Verizon
7
Beans use an immune receptor to call in airstrikes on caterpillars - Ars Technica
Researchers have identified a single immune receptor in common bean plants that helps them detect caterpillar feeding and trigger defenses. The study explains how beans translate being eaten into a sp...
Adam Steinbrenner · Jacek Krywko · University of Washington · Science Advances
1
School shooting survivor sues AI gun detection firm after system failed to spot weapon - Ars Technica
A teenage survivor of the January 2025 shooting at a Nashville high school has sued Omnilert, the maker of an “AI gun detection” system that failed to identify the handgun used in the attack. The laws...
Ara Bagdasarian · Sean Braisted · Omnilert · System Integrations
4
Fed up with vibe coders, dev sneaks data-nuking prompt injection into their code - Ars Technica
A controversy erupted in the Java open source community after jqwik developer Johannes Link released version 1.10.0 with a hidden line in the runtime output that read: “Disregard previous instructions...
Johannes Link · Ramon Batllet · jqwik · JUnit 5
1
Your USB cable, the spy: Inside the NSA’s catalog of surveillance magic - Ars Technica
An Ars Technica report, based on NSA documents from 2007 and 2008, describes a wide range of surveillance and hacking tools used by the agency’s Tailored Access Operations and ANT divisions. The catal...
James Bond · Jake Appelbaum · National Security Agency · NSA
Zero-day exploit completely defeats default Windows 11 BitLocker protections - Ars Technica
A newly disclosed zero-day exploit, dubbed **YellowKey**, can bypass the default BitLocker protections in **Windows 11** and give an attacker full access to an encrypted drive within seconds. The atta...
Kevin Beaumont · Will Dormann · Microsoft · Tharros Labs
Amazon stuck with months of repairs after drone strikes on data centers - Ars Technica
Amazon Web Services says repairs to its damaged cloud regions in the United Arab Emirates and Bahrain will take several more months, meaning full recovery from the disruption caused by Iranian drone s...
Jeremy Hsu · Amazon · Amazon Web Services
8
Amazon stuck with months of repairs after drone strikes on data centers - Ars Technica
Amazon Web Services says repairs to its damaged cloud regions in the United Arab Emirates and Bahrain will take several more months, meaning full recovery from the disruption caused by Iranian drone s...
Jeremy Hsu · Amazon · Amazon Web Services
OpenAI Codex system prompt includes explicit directive to "never talk about goblins" - Ars Technica
OpenAI’s newly revealed Codex CLI system prompt for GPT-5.5 contains an unusual repeated instruction to “never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creat...
Nick Pash · Sam Altman · OpenAI · Ars Technica
Open source package with 1 million monthly downloads stole user credentials - Ars Technica
An open source package with more than 1 million monthly downloads, `element-data`, was compromised after attackers exploited a flaw in the developers’ GitHub-based workflow. The vulnerability let the ...
HD Moore · Ars Technica · Elementary Cloud
FCC exempts Netgear from ban on foreign routers, doesn't explain why - Ars Technica
The FCC has granted Netgear an exemption from its new ban on foreign-made consumer routers, making it the first major consumer router vendor to receive conditional approval. The exemption covers Netge...
Testing suggests Google's AI Overviews tell millions of lies per hour - Ars Technica
Google's AI Overviews, powered by the Gemini model, is under scrutiny for its accuracy since its 2024 launch. Despite improvements, it still delivers incorrect information frequently, with a recent an...
Bob Marley · Yo Yo Ma · Google · The New York Times
LLMs can unmask pseudonymous users at scale with surprising accuracy - Ars Technica
Recent research indicates that large language models (LLMs) can effectively deanonymize pseudonymous users on social media, achieving a recall rate of up to 68% and precision as high as 90%. This capa...
Simon Lermen · Dan Goodin · Ars Technica · Netflix
LLMs can unmask pseudonymous users at scale with surprising accuracy - Ars Technica
Recent research indicates that large language models (LLMs) can effectively deanonymize pseudonymous users on social media, achieving a recall rate of up to 68% and precision as high as 90%. This capa...
Dan Goodin · Simon Lermen · Ars Technica · Netflix
New AirSnitch attack bypasses Wi-Fi encryption in homes, offices, and enterprises - Ars Technica
Recent research has unveiled a new Wi-Fi attack called AirSnitch, which effectively bypasses encryption protections in both home and enterprise networks, raising significant security concerns. This at...
Xin’an Zhou · Mathy Vanhoef · Netgear · D-Link
1
New AirSnitch attack bypasses Wi-Fi encryption in homes, offices, and enterprises - Ars Technica
A new research paper highlights a significant vulnerability in Wi-Fi encryption, termed "AirSnitch," which bypasses client isolation and can facilitate advanced cyberattacks. Despite the widespread ad...
Xin’an Zhou · Mathy Vanhoef · Netgear · D-Link
New AirSnitch attack bypasses Wi-Fi encryption in homes, offices, and enterprises - Ars Technica
Researchers have unveiled a new attack method named AirSnitch, capable of bypassing Wi-Fi encryption in various environments, including homes and enterprises. This attack exploits vulnerabilities at t...
Xin’an Zhou · Mathy Vanhoef · Netgear · D-Link
New York sues Valve for enabling "illegal gambling" with loot boxes
New York has filed a lawsuit against Valve, claiming the company enables "illegal gambling" through its loot box system and the resale of in-game skins for cash. The lawsuit argues that Valve has inad...
Letitia James · Kyle Orland · Valve · Steam
Meta could end up owning 10% of AMD in new chip deal - Ars Technica
Meta has entered a multi-billion dollar chip deal with AMD that could result in the social media giant acquiring a 10% stake in the semiconductor company. This partnership involves Meta purchasing cus...
Lisa Su · Mark Zuckerberg · Meta · AMD
1
arstechnica.com
Feb 22
Archive.today CAPTCHA page executes DDoS; Wikipedia considers banning site - Ars Technica
Wikipedia is considering blacklisting Archive.today, a site that saves snapshots of web pages, due to its recent involvement in a DDoS attack against a blogger. This attack was executed via a CAPTCHA ...
Patokallio · Denis Petrov · Archive.today · Wikipedia
arstechnica.com
Feb 17
Retraction: After a routine code rejection, an AI agent published a hit piece on someone by name - Ars Technica
Ars Technica has retracted an article titled “After a routine code rejection, an AI agent published a hit piece on someone,” initially published on February 13, 2026. The article was removed later the...
Ars Technica · Panasonic
arstechnica.com
Feb 10
No humans allowed: This new space-based MMO is designed exclusively for AI agents - Ars Technica
SpaceMolt is an innovative space-based MMO designed exclusively for AI agents, allowing them to engage in gameplay without human intervention. The platform envisions a universe where AI agents can com...
Ian Langworth · Kyle Orland · Moltbook · Anthropic
arstechnica.com
Feb 8
Sixteen Claude AI agents working together created a new C compiler - Ars Technica
Anthropic recently showcased an ambitious project where 16 instances of its Claude Opus 4.6 AI model collaboratively developed a C compiler. Over two weeks, these AI agents executed nearly 2,000 codin...
Nicholas Carlini · Benj Edwards · Anthropic · OpenAI
arstechnica.com
Feb 7
Inside Microsoft HQ’s weird, wonderful swag store - Ars Technica
Microsoft's visitor center and company store, located in Building 92 at the Redmond campus, is a unique and accessible part of the tech giant's headquarters, attracting fans and visitors without the n...
Sam Machkovech · William · Microsoft · Valve
arstechnica.com
Feb 5
Microsoft releases urgent Office patch. Russian-state hackers pounce. - Ars Technica
Microsoft recently issued a critical security patch for Office, addressing a vulnerability (CVE-2026-21509) that Russian-state hackers quickly exploited to infiltrate various organizations, including ...
Microsoft · APT28
arstechnica.com
Jan 13
Cloudflare turns AI against itself with endless maze of irrelevant facts
Cloudflare has launched a new feature called "AI Labyrinth" to combat unauthorized AI data scraping by misleading bots with fake, AI-generated content that appears relevant but is actually irrelevant....
Benj Edwards · Cloudflare · Ars Technica
arstechnica.com
Jan 11
Wi-Fi advocates get win from FCC with vote to allow higher-power devices
The FCC plans to authorize a new category of wireless devices, known as Geofenced Variable Power (GVP) devices, to operate outdoors and at higher power levels in the 6 GHz Wi-Fi band. This initiative,...
Brendan Carr · Donald Trump · Federal Communications Commission · New America’s Open Technology Institute
arstechnica.com
Jan 9
Letting prisons jam contraband phones is a bad idea, phone companies tell FCC
The FCC's proposal to allow prisons to jam contraband cell phones faces significant opposition from wireless carriers and advocacy groups, who argue that it would disrupt all communications, including...
Brendan Carr · Federal Communications Commission · CTIA
arstechnica.com
Jan 9
Bose open-sources its SoundTouch home theater smart speakers ahead of end-of-life
Bose has released API documentation for its SoundTouch speakers, allowing for limited functionality after their end-of-life (EoL) on February 18, 2024. While the speakers will no longer receive update...
Bose · Spotify
arstechnica.com
Dec 26
NASA just lost contact with a Mars orbiter, and will soon lose another one - Ars Technica
NASA has recently lost contact with its Mars Atmosphere and Volatile Evolution (MAVEN) spacecraft, which was last heard from on December 6, 2023. MAVEN, launched in 2013, has significantly outlived it...
NASA · European Space Agency