The Register
Coverage of The Register in the Nexus archive.
- C programmers commit fresh crimes against readability
The article criticizes C programmers for writing code that sacrifices readability, labeling their practices as 'crimes against readability.' It is published in The Register's offbeat section and has minimal community engagement with zero comments.
- User swore hacker called General Failure had invaded his PC
A company vice president reported a hacker named 'General Failure' was accessing his files, but a sysadmin clarified the error message indicated a disk failure. The sysadmin, Lee, resolved the issue by diagnosing hardware problems instead of a security breach.
- Portuguese restaurant kiosk software gives Windows indigestion
A Portuguese restaurant chain's kiosk software, WinRestKioskWPF.exe, is failing on Windows 10/8.x due to a missing digital signature, displaying a warning dialog that disrupts customer use. The software, part of the WinRest Point of Sale suite, is described as 'ultra-secure and affordable' but is causing operational issues.
- Sysadmin broke hardware worth more than he made in a month – and lied his way out of the mess
Hank, a sysadmin, accidentally dropped an expensive 750MB SCSI drive, lied to his boss about it malfunctioning, and successfully replaced it with a new drive. The incident led to him learning to handle hardware carefully.
- Portuguese bank sign's storage is about to cash out
A digital sign at Banco CTT in Lisbon displays a 'S.M.A.R.T. Status Bad' error, indicating potential storage failure. The sign, showing 4GB of installed memory, has sparked humor about its value amid rising RAM prices. The article notes Lisbon's attractions while urging the bank to address the hardware issue.
- How dare you stop data loss – that's not your job!
Terry, a municipal IT worker in the 1980s, faced reprimands for both clocking irregular hours and attempting to fix a data-loss issue caused by a flawed MOVE command on an IBM mainframe. His efforts to prevent data loss by rewriting the script were dismissed as overstepping his job responsibilities, leading to his eventual departure.
- Users claimed they’d never seen a spell checker and panicked at the sight of red squiggles
A company's use of an outdated MS Word document for recording orders led to confusion when a new employee, Mitch, panicked over red underlines caused by Word's spell checker. Cooper, a staff member, explained that the red lines flagged SKUs as non-dictionary words and suggested disabling the spell checker or ignoring the warnings.
- Qualcomm said to be circling AI chip biz Tenstorrent in $10B RISC-V power play
Qualcomm is reportedly in talks to acquire AI chip company Tenstorrent, a RISC-V-based firm, in a deal valued at $8 billion to $10 billion. The potential acquisition aligns with Qualcomm's datacenter ambitions and efforts to expand its RISC-V ecosystem, following its earlier purchase of Ventana Micro Systems. Tenstorrent's CEO, Jim Keller, is known for his work at AMD, Apple, and DEC.
- This is your BIOS speaking. Please fix me. Your PC is broken
A university IT technician named Jackson encountered a PC that rebooted and played a Chinese voice message during a Power-On Self-Test (POST). The issue stemmed from a 'talking error BIOS' feature on the motherboard, which defaulted to Chinese after a dead CMOS battery reset settings. Jackson resolved it by changing the BIOS language to English, revealing the error was related to a missing floppy drive.
- Next stop, C:\ ... Paris Metro screen goes off the tracks
A Paris Metro screen on Line 11 displayed a directory browser instead of transit information, spotted at the Chatêlet terminus in June. The incident, humorously compared to France's can-can dance, may stem from technical issues, heatwave stress, or fan enthusiasm following PSG's Champions League victory. Line 11, known for its rubber-tired trains and historical ties to the Belleville funicular, was recently extended.
- Consultant mistakenly deleted a ton of data – but reported it as a bug
A test automation consultant accidentally deleted data using a script, reported it as a bug, and the client's support team restored it, taking responsibility for the incident. The consultant avoided blame as the client attributed the error to a faulty SaaS script.
- Our systems editor flew all the way to Taiwan and still couldn't get away from AI
The Register's systems editor Tobias Mann reported from Computex 2026 in Taipei, highlighting how AI is driving chipmakers to prioritize demand for AI hardware over other products. Key announcements included Nvidia's N1X high-end notebook SOC for Windows and Intel's updates, though the article questions whether rising hardware costs will stabilize.
- Yet another Cisco SD-WAN 0-day under attack, and no patch in sight
Cisco's SD-WAN management software is under attack due to a high-severity zero-day vulnerability (CVE-2026-20245), which allows authenticated attackers to escalate privileges and execute commands. Cisco has not yet released a patch, and this is the sixth SD-WAN vulnerability exploited since the year began.
- Tech support chap hauled out to help SWAT team saw his life flash before his eyes
A tech support worker named Solomon was unexpectedly called to assist a sheriff's office raid, where he encountered a high-stakes situation involving a SWAT team. Despite initial fears, he was later informed the operation didn't require his IT skills, and he celebrated with a meal afterward.
- Commvault says it's time to rethink resiliency as AI crooks leave victims in a 'dark, dead' state
AI-enabled cybercriminals are using advanced tools to destroy virtual machines and hypervisors, leaving infrastructure in a 'dark, dead' state. Commvault's CTO emphasizes the need for organizations to adopt strategies beyond backups, including air-gapping and pressure-testing recovery plans, to counter evolving AI-driven threats.
- Claude celebrates Anthropic's stock market float with blockbuster ... outage
Claude, a chatbot and coding tool developed by Anthropic, experienced an outage following the company's IPO filing. Anthropic, valued at $965 billion, faces challenges as users reported disruptions, while competitors like OpenAI and SpaceX also anticipate major IPOs.
- Northern Ireland cops issue PSA after official phone number spoofed by scammers
The Police Service of Northern Ireland (PSNI) warned the public about scammers spoofing its switchboard number to impersonate police and solicit gift cards and bank details. A separate case involved an elderly woman defrauded of over £250,000 through a cryptocurrency scam involving malware.
- Password manager Dashlane suspends customer accounts amid brute-force attacks
Password manager Dashlane suspended customer accounts following brute-force attacks, which began on Sunday and led to unauthorized login attempts from Korea and Russia. The company restored accounts after investigating the incident but later changed the status to 'monitoring.' Dashlane confirmed no internal systems were compromised, though users criticized the lack of public communication and faced issues with two-factor authentication.
- LLMs Are Closer to Religion Than They Appear
The article argues that Large Language Models (LLMs) share similarities with religion, cautioning against those who prefer this analogy. It highlights a discussion around the implications of framing AI in religious terms.
- Techie expensed a bag of oranges and then juiced up a stupid security incident
A reader named Kirk worked at a government building swapping out computers. His colleague expensed a bag of oranges and threw them from a window, causing a security response. The incident led to a lesson about perception but no consequences for Kirk or his colleague.
- ShinyHunters adds Charter to trophy shelf after 4.9M customer records leak
ShinyHunters leaked 4.9 million Charter Communications customer records, including names, email addresses, phone numbers, and physical addresses, after the company allegedly refused their extortion demands. Charter confirmed an investigation but claimed no sensitive data was exfiltrated, while the breach is part of a broader pattern of ShinyHunters targeting organizations, including a recent Carnival Corporation data theft.
- ReactOS brings its Windows NT tribute act to ARM64
ReactOS, an open-source project recreating Windows NT, has achieved a milestone by booting on ARM64 hardware, including the Raspberry Pi 5. The build is experimental, with limited functionality, but marks a significant technical feat after eight months of work. The project aims for Windows Server 2003 compatibility and celebrates its 30th anniversary.
- Bare metal cloud servers now cheaper and more readily available than on-prem hardware, says Nutanix CEO
Nutanix CEO Rajiv Ramaswami claims cloud bare metal servers are now cheaper and easier to acquire than on-prem hardware due to hyperscalers' bulk purchasing power. He highlights persistent high memory/storage prices, growing on-prem AI adoption for cost predictability, and Nutanix's Q3 2026 results showing 730 new clients and $703 million revenue. The company shifted to support external storage, securing deals with Everpure and Dell.
- InPost locker caught shipping unactivated Windows
InPost lockers in the UK are using unactivated Windows operating systems, causing activation warnings. The issue may stem from hardware changes or lack of updates, with potential risks to system stability.
- Logitech unveils a cushioned mouse for all-day use
Logitech has released the Signature Comfort Plus M850 L mouse and MK880 keyboard, both designed for right-handed users. The mouse features a cushioned design, Bluetooth connectivity, and recycled plastic components, while the keyboard offers an ergonomic curve and quiet keys. However, the lack of left-handed support and non-removable palm cushions are noted drawbacks.
- Under-trained techie didn't claim overtime for mistakenly failing to phone it in
April, an entry-level IT technician with CompTIA A+ certification, mistakenly connected both Ethernet ports of a VoIP phone during a clinic deployment, causing a complete internet outage. Despite resolving the issue, she avoided claiming overtime, believing she didn't deserve it for the error. The team responded calmly, fixing the problem quickly.
- Techie claims Trump Mobile website was leaking thousands of people's data
A security researcher identified a critical vulnerability on Trump Mobile's website that exposed personal data of approximately 27,000 customers through a simple HTTP POST request. The flaw allowed access to names, addresses, email addresses, phone numbers, and account numbers before being patched. The researcher disclosed findings to YouTube creators after receiving no response from Trump Mobile.
- Threat hunters find Google API keys still usable 23 minutes after deletion
Security researchers at Aikido discovered that Google API keys remain usable for up to 23 minutes after deletion, creating a significant vulnerability window for attackers. When combined with Google's automatic billing tier upgrades, compromised keys can lead to devastating financial charges as bad actors exploit Gemini and other services before the credentials are fully revoked across Google's infrastructure.
- HackerOne takes an axe to its bug bounty rewards
HackerOne has drastically reduced bug bounty rewards through its Internet Bug Bounty (IBB) program, cutting payouts by 68-88% across all severity levels. The program remains paused while the platform evaluates adjustments, leaving security researchers waiting for payments on previously submitted vulnerabilities.
- Microsoft storms RAMPART, adds Clarity to agentic AI safety
Microsoft open-sourced two AI safety tools: RAMPART, a pytest framework for red-teaming agentic AI applications in CI/CD pipelines, and Clarity, an agent that helps teams evaluate design decisions before implementation. RAMPART enables developers to simulate attack scenarios and verify AI agents stay within approved boundaries, while Clarity acts as a structured advisor providing questions similar to experienced architects and safety engineers.
- UK’s Education Committee: Social media ban a must to save children’s mental health
The UK Parliament's Education Committee is advocating for a statutory ban on social media for under-16s, citing serious harms to children's mental health and development caused by addictive platform design elements. The committee also calls for broader regulatory frameworks covering gaming, messaging services, and AI platforms. This position was communicated to the UK government as part of an ongoing consultation that began in March.
- Zombie user account let hackers control the city’s water
A municipality's water utility was compromised by hackers who gained access through a dormant employee account belonging to a former auditor named Greg. The attacker exploited a leaked password that Greg had reused across work and personal accounts, gaining extensive privileges including domain admin and SCADA operator access. The incident highlights critical failures in account management and the importance of regular access audits.
- Google accused of pushing 'free for life' G Suite users onto paid plans
Google is warning long-time G Suite Legacy users to start paying for Workspace subscriptions or lose access to core services, citing commercial use. Users are being pushed towards paid plans despite claiming personal use. The crackdown has sparked complaints on Reddit.
- Linus Torvalds says Linux security list is becoming ‘unmanageable’ due to AI bug reports
Linux founder Linus Torvalds stated that the Linux security list is becoming unmanageable due to a flood of AI-generated bug reports, resulting in significant duplication. This issue does not apply to specific exploits like the Copy Fail exploit, which was detected with AI assistance. The Linux security list is struggling with manageability.
- Linux security mailing list 'almost unmanageable'
Linux security mailing list has become almost unmanageable due to AI-powered bug hunters, according to Linus Torvalds. The Linux security mailing list is struggling to keep up with the volume of reports. This has led to concerns about the effectiveness of the list in addressing security issues.
- Backup script ingested an accidental asterisk and deleted everything
A 21-year-old programmer, Miller, wrote a backup script that accidentally deleted all files and data due to an unexpected asterisk character. The incident occurred in 1981 on a mainframe system where Miller worked. He was able to restore the data after a day of work.
- Linus Torvalds says AI-powered bug hunters have made Linux security mailing list ‘almost entirely unmanageable’
Linus Torvalds stated that the Linux security mailing list has become almost entirely unmanageable due to duplicate bug reports from AI-powered tools. He emphasized the need for productive use of AI tools and adding real value beyond automated reports. The situation highlights the challenges of leveraging AI in software security.
- Surprise AI bills leave AWS and Google Cloud users aghast
AWS and Google Cloud users are receiving unexpected bills in the tens of thousands of dollars due to AI services usage, with providers showing little urgency in resolving the matter. The issue is attributed to compromised API keys and a policy change by Google. Users are advised to be cautious when using AI services on cloud platforms.
- AI-generated code is 'pain waiting to happen'
Moshe Sambol, VP of customer solutions at Lightrun, discusses the challenges of adopting AI tools in software development, citing the gap between business expectations and developers' abilities to effectively use these tools. Many developers are not adequately trained to work with AI-generated code, leading to potential bugs and errors. The lack of understanding and context provided by generative AI systems exacerbates this issue.
- Cloud-managed earbuds sound strange - as a concept, and on a plane
Dell's cloud-managed earbuds have poor active noise cancelling and connectivity issues, but offer firmware updates and management tools through the Device Management Console. The earbuds are priced higher than Apple's AirPods and may not be worth the cost for personal use. The Device Management Console is primarily intended for managing docks and displays.