A cyberattack claimed by pro-Iran hackers has caused a “global network disruption” to a major US medical device maker, according to a company statement.
Gemini 3.1 Pro scores 77.1% on ARC-AGI-2. Researchers called that benchmark unsolvable two years ago. The labs solved it. They will not stop here. The question is not whether AI keeps advancing. It is who captures the value when it does.
Standard Kernel, which is making AI-powered GPU optimization software, raised a $20M seed led by Jump Capital, with General Catalyst and others participating (Clinton Nwachukwu/Ventureburn)
"We see a future where intelligence is a utility like electricity or water and people buy it from us on a meter and use it for whatever they want to use it for" Sam Altman, of OpenAI, the company behind ChatGPT, told the BlackRock Infrastructure Summit in D.C.
How to Make a File Encryption Tool That Self-Destructs the Key (High-Security Design Guide)
In cybersecurity, deleting a file isn’t enough. Even formatting a drive isn’t enough. If someone gains access to storage, backups, or forensic recovery tools, deleted files can sometimes be recovered.…
AI *can* help predict wildfire spread, but UB researchers found it’s not a complete solution yet. Deep learning needs pairing with existing physics-based models. 🔥 #wildfires
There are many smart home devices that can compromise the security of your personal and banking information. These include thermostats and baby monitors.
Tech giants are currently strip-mining the power grid to teach chatbots how to hallucinate more convincingly. While OpenAI and xAI cosplay as environmentalists, they are actually just firing up vintage coal plants and gas turbines because waiting for "clean energy" is bad for the quarterly alpha.
Microsoft to Block Windows 11 and Server 2025 Automated Installation After Critical RCE Vulnerability Microsoft has announced a two-phase plan to disable the hands-free deployment feature in Window...
Alerts from NWS Shreveport LA (UNOFFICIAL)@shv.nws-bot.us
Felsenthal [Union Co, AR] Law Enforcement reports Tstm Wnd Dmg at 15 Mar, 7:22 PM CDT -- A tree was blown down across a road in the Felsenthal community.
at Mon, 16 Mar 2026 03:31:10 +0000 via IEMbot Additional Details Here.
What Is Threatening IBM i Security Now The nature of cybersecurity threats is always changing, which requires constant vigilance by those who value security. IBM i is just as exposed to threats as ...
Your phone's "off" switch was never a lock.
CVE-2025-20435: 875M Android devices.
USB in. 60 seconds. PIN cracked, encryption stripped, before the OS even blinks.
Check your MediaTek chip. Patch now. Or hand-deliver your secrets. www.forbes.com/sites...
Google's and Accel's joint AI startup accelerator Atoms picked five Indian startups from 4K+ applications, saying ~70% of rejected applicants were AI "wrappers" (Jagmeet Singh/TechCrunch)
Microsoft Defender for Endpoint is getting a new security recommendations that will help organizations boost their security posture in an increasingly dangerous world. #Microsoft#Defender
Sources: Microsoft sidelined some Copilot-branded AI features on Windows 11 after the Recall delay, and shipped some promised features without the Copilot name (Zac Bowden/Windows Central)
The latest update for #CatoNetworks includes "Cato CTRL Threat Brief: Middle East Escalation and Summary of Notable Iranian-Linked CVEs" and "When the M&A Deal Closes, Is Your Architecture Accelerating Time to Value?".
The latest update for #KnowBe4 includes "How do I Send a Secure Email in Outlook?" and "Received Someone Else's Confidential Email? Here's What To Do.".
"Cybercriminals have successfully infiltrated a critical intelligence infrastructure platform, compromising sensitive operational data and contractor information used by the Central Intelligence Agency and other federal agencies." cyberpress.org/cia-and-othe...
Microsoft Tracks Storm-2561 In Fake VPN Client Credential Theft Scheme Microsoft says a cybercriminal group it tracks as Storm-2561 is running a credential theft campaign that uses fake VPN clients...
OpenAI CEO Sam Altman admits AI is disrupting capitalism by eroding worker power. He acknowledges AI lets businesses undermine workers as it shifts the economy to one of abundance. #AI #AI#News
@rbreich.bsky.social Dementia. Trump duplicates the Vietnam war thinking of General Curtis E. LeMay "We’re going to bomb them back into the Stone Age". The US lost that war 50 years ago.
Meta Plans to Remove End-to-End Encryption From Instagram Direct Messages Permanently Meta will remove end-to-end encryption (E2EE) from Instagram Direct Messages on May 8, 2026, ending the platfor...
A new debate questions whether Bitcoin miners pivoting to artificial intelligence could threaten network security. Crypto trader Ran Neuner warns that AI’s higher electricity payments are drawing miners away, potentially weakening Bitcoin. Others, like Bitcoin pioneer Adam Back, argue the network’s…
OpenAI's Sora Video Generator Set to Join ChatGPT SummaryOpenAI reportedly plans to integrate its Sora video generation model directly into ChatGPTThe move aims to revive interest after the sta...
The 🛡️ Straylight Sentinel Intelligence Report for Monday, March 16, 2026 | 05:23 UTC is ready for all of my #cybersecurity and #infosec friends out there. Good Morning Europe!
Just saw an interesting comment recently from Dario Amodei about how he spends his time running Anthropic. He said around 30–40% of his time goes into company culture. The real focus becomes making sure the mission is clear, and everyone is pulling in the same direction.
Microsoft Releases Emergency Patch for Critical RRAS RCE Flaw in Windows 11 Microsoft has released an urgent out‑of‑band security update to address several critical vulnerabilities in the Windo...
“Iran pulled off likely the most significant wartime cyberattack against the U.S. in history, leveraging its hacking powers to cause major disruptions at a global medical-equipment firm that struggled to bring itself back online in recent days.” Gift
OpenAI's Sora Video Generator Set to Join ChatGPT SummaryOpenAI reportedly plans to integrate its Sora video generation model directly into ChatGPTThe move aims to revive interest after the sta...
Researchers at Budapest University of Technology linked compare-exchange (sorting) networks to QUBO, reducing permutation encoding from n² to O(n log₂ n) binary variables — enabling unbiased sampling, multiplication & inversion for cryptography and combinatorial design.
As AI models commodify, Apple's plans to spend just $14B on 2026 capex, far below hyperscalers' combined $650B, or 90%+ of their cash flow, may be a genius move (Horace Dediu/Asymco)
Google’s Gemini AI chatbot finally rolling out to all Hongkongers Google has said it will gradually open its Gemini Artificial Intelligence (AI) tool to all users in Hong Kong, with the accompany...
Google’s Gemini AI chatbot finally rolling out to all Hongkongers Google has said it will gradually open its Gemini Artificial Intelligence (AI) tool to all users in Hong Kong, with the accompany...
14,000 Routers are infected by Malware, named KadNap, that have been conscripted into a proxy Network that anonymously carries Traffic used for Cybercrime - Technical Report by the Black Lotus Labs team at Lumen #Malware#Infosecblog.lumen.com/silence-of-t...
Meta's answer to Gemini 3.0 reportedly delayed due to performance issues tech.yahoo.com/ai/br...
The launch of the new model, code-named Avocado, may miss its reported targeted release date of mid-March.
www.forrester.com/blogs/what-i... OpenAI has culled its instant checkout option with partner Shopify. Although small numbers (35% in biggest GenZ segment) use 'answer engines' building trust and getting transaction and inventory management in sync between sites and chatbots proving too challenging.
Boston Metro Weather Forecast Bot@boston-4castbot.bsky.social
NWS Boston (via Facebook): [Heavy rain and Wind] Widespread showers and embedded t-storms through tonight will bring heavy rainfall and possible flooding. Also, a period of strong to damaging winds expected tonight with gusts 45-65 mph. Flood Watches, High Wind Warnings and Wind Adviso… [Link]#MaWX
Parrot SA will integrate SEALSQ's post-quantum cryptography semiconductors into its next-generation commercial micro-UAVs, advancing quantum-resistant security for drone hardware announced at Embedded World 2026.
Google, Microsoft, LinkedIn, Meta, Amazon, OpenAI, Adobe, and Match sign a voluntary pledge to share threat intelligence on scammers abusing their services (Sam Sabin/Axios)
Suggestions for improving the ChatGPT mobile app Hello OpenAI Support Team, I hope you are doing well. I would like to share a couple of suggestions that could significantly improve the user experi...
Security shouldn’t slow down your sprint. In this article, Christian Schneider shows how attack trees + 45-minute threat huddles can help teams identify risks early and turn them into actionable security tasks.
Strengthen your digital security: review app permissions on your phone. Are they still needed? Limiting access enhances privacy. Follow NIST's Cybersecurity Framework principles: secure what you can! ✨ Source: https://www.nist.gov/cyberframework#Privacy#Security
There’s been a row between Anthropic and Hesgeth about the use of Anthropic’s Claude system to identify people targets in Iran. If human judgement is left out of the loop accountability is gone.
eSentire's TRU team investigated an open-directory web server attributed to the Iranian state-sponsored MuddyWater group and identified a file that deploys the Tsundere botnet, a suspected malware-as-a-service (MaaS) offering of Russian origin. www.esentire.com/blog/muddywa...
The TechBeat: Google Gemini vs Anthropic Claude vs OpenAI ChatGPT vs xAI Grok: The Ultimate Comparison (3/15/2026)
HackerNoon's Techbeat features trending tech stories covering various aspects of technology. One article explores the best suits from Marvel's Spider-Man. Anothe… #chatgpt#gemini#gpt
Mastering core network protocols like TCP/IP, HTTP, SMTP, and FTP builds a strong foundation for understanding secure communication and defending against web and email attacks. #NetworkBasics#SecureEmail#USA
IBM X-Force researchers discovered a likely AI-generated malware which they are dubbing Slopoly, used during a ransomware attack. The operators are part of a group tracked as Hive0163, whose main objective is extortion through large-scale data exfiltration and ransomware. www.ibm.com/think/x-forc...
In this week’s news: Cyberattackers Don't Care About Good Causes, Red-teamers unleash AI agent on McKinsey’s chatbot, gain full access in two hours buttondown.com/BagheeraAlte...
Samsung Galaxy S26 Ultra review: Privacy Display works well, meaningful camera upgrades, and sleeker than S25 Ultra, but still huge, expensive, and lacks Qi2 (Allison Johnson/The Verge)
WhiteBridge AI raises $3M to redefine standards in people search and research WhiteBridge AI, a team developing anAI-powered people search and research engine, has raised a $3M seed round. Theround...
It's interesting that I don't see more reports on how Anthropic's *lifetime* revenue (up to the start of March 2026) is a bit over $5 billion: www.wheresyoured.at/the-beginnin...
Attackers are exploiting AI faster than defenders can keep up, new report warns Cybersecurity is entering “a new phase” as artificial intelligence tools have matured and given IT defenders sign...
Foxconn reports Q4 revenue up 22% YoY to ~$81B, net profit down 2% YoY to ~$1.4B, below ~$1.9B est., due to a higher tax expense, and a 5.9% gross profit margin (Sherry Qin/Wall Street Journal)
Amsterdam-based Nebius says Meta plans to spend up to $27B over five years for access to AI infrastructure, starting with $12B of capacity in early 2027 (Kurt Wagner/Bloomberg)
Anthropic’s new Code Review feature uses multiple AI agents to analyze pull requests and flag potential bugs. However, some developers question whether AI systems reviewing code written by other AI tools makes sense at all. tessl.io/blog/anthrop...
Google Looker Studio Vulnerabilities Enable Attackers to Exfiltrate Data from Google Services Google Looker Studio was affected by nine high‑impact “LeakyLooker” vulnerabilities that could ha...
Memo: Alibaba unveils Alibaba Token Hub, a new AI unit that consolidates research, consumer-facing apps, AI products, and more, as it focuses on monetizing AI (Luz Ding/Bloomberg)
𝟭𝟰 + 𝟭𝟱 𝗔𝗽𝗿𝗶𝗹: 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗖𝗼𝗺𝗽𝘂𝘁𝗶𝗻𝗴 𝗶𝗻 𝗝𝘂𝗹𝗶𝗮: an introduction to deep learning for researchers who are familiar with the basics of (non-deep) machine learning.
Registration is now open! www.esciencecenter.nl/event/effici...
Week in review: AiTM phishing kit used to hijack AWS accounts, year-long malware campaign targets HR
Here’s an overview of some of last week’s most interesting news, articles, interviews and videos: Turning expertise into opportunity for women in cybersecurity Speaker diversity i… #hackernews#news
GovAI and Brookings: many of the people most at risk if AI transforms work are also the best placed to find new jobs, as researchers disagree over AI's impact (Washington Post)
Attackers are now padding phishing emails with legitimate links and hundreds of blank lines to confuse AI detection. The malicious bit sits at the top. The noise sits below. Your security tools are squinting at a hay...
“…OpenAI is struggling to lift ChatGPT’s restrictions on NSFW content while keeping more harmful scenarios off limits, such as those depicting nonconsensual behavior or child sexual abuse.”
On Friday... Companies House was made aware of a security issue which meant that a logged-in user of our WebFiling service could potentially access and change some elements of another company’s details without their consent after performing a specific set of actions.
Russian Hackers Are Going After Signal Accounts, Dutch Officials Warn
Signal and WhatsApp offer end-to-end encryption, but Dutch officials say Russian hackers attack user accounts in two ways.
- by acting as customer support asking for your pin & verification code
- by getting you to use a QR Code
Signs suggest Iranian-linked #cyber groups may be shifting toward disruptive ops - probing #APIs & apps that power Western infrastructure. Early reconnaissance often precedes bigger #attacks. Stay vigilant: monitor traffic & secure access.
Android 17 Advanced Protection Mode to Block Malicious Service Usage Google is preparing to launch Android 17, bringing a comprehensive set of new APIs and system capabilities to fundamentally impr...
🚨 In this week’s threat alert, CrowdSec reports on CVE-2026-23744, a critical RCE in MCPJam Inspector. Exploitation attempts are rising, targeting exposed dev environments.
I. P. De Simeone, G. Maltese, V. Cambier, J-P. Likforman, M. Ravaro, L. Guidoni, F. Baboux, S. Ducci
Deep-Learning-Designed AlGaAs Interface Linking Trapped Ions to Telecom Quantum Networks https://arxiv.org/abs/2603.12846
How AI Agents Decide Which Brands To Recommend: Trust Is The New Ranking Factor Discover how AI agents decide which brands to recommend and why trust is essential in the new marketing landscape. Th...
But that doesn't mean you're visible.
Google. AI answers. Trust signals.
If one of them breaks, customers pick someone else.
That’s exactly what Evident checks.
A Register reader told us he tested the LLM via Microsoft's platform using Startup Credits, however, this ended with credit card charges. Microsoft and Anthropic then each directed our reader to contact the other company in order to get a resolution. www.theregister.com/2026/03/13/a...
Microsoft Locks Down Discord Server After People Wouldn’t Stop Making Fun Of AI ‘Microslop’
We've noted how Microsoft is a little sensitive about AI slop at the moment. Back in January, CEO Satya Nadella wrote a well-circulated blog post lamenting critics of "AI slop" and demanding the public…
Storing payment details in your browser or #online#shops is convenient but poses a high #security risk. Read on to find out what you should do instead.
OK, I just spotted Bigfoot :) I finally started seeing ads this morning in ChatGPT, and a number of them. So OpenAI is definitely ramping up ads... They aren't particularly exciting and some weren't really targeted, but hey, at least I finally spotted some.
Forget stolen credentials and misconfigurations; AI means vulnerability exploits that beat patching cycles are the top cause of compromises in the cloud. www.darkreading.com/cloud-securi...
OpenAI’s adult mode will reportedly be smutty, not pornographic – ChatGPT’s “adult mode” is expected to support saucy text conversations at launch, but not images, voice, or video. OpenAI said earlier this month that it was delaying the rollout to focus on hi... https://tinyurl.com/23d76yps#Chatbot
"How Anthropic may benefit from its fight with Trump.
Anthropic’s high-stakes gamble to take on the Trump administration could give it an advantage in the AI race."
As companies deepen their #Scope3 accounting and map environmental impacts across supplier tiers, they're also creating new #cybersecurity vulnerabilities.
The same visibility work that enables sustainability reporting can expose organizations to #cyberphysical security risk.
Gemini CLI plan mode: read-only exploration as the default.
Navigate your codebase. Map dependencies. Ask clarifying questions. Propose a plan. Implement only when approved.
Now enabled by default for all users devops.com/gemini-cli-p...#DevOps#GeminiCLI#AIagents#PlanMode
I love the Hacking Humans podcast which is about social engineering done with technology. This is a great episode about how much info is really contained in Metadata (which the tech co’s tell us is not a big deal)
“In proposing to allow sexually explicit conversations with its popular chatbot, OpenAI exposed fractures over how to balance rapid user growth and digital freedom with safety and child protection.”
