Data-Leaking Ransomware Report - IT & Communications 2025 Welcome to a special DLR Report solely focused on the IT and Communications Sector, an exclusive presentation of Data-Leaking Ransomwar...
Using Claude AI in military operations, especially after the directive to stop, likely constitutes a violation of Anthropic's Terms of Service. www.cbsnews.com/news/anthrop...
"This data broker industry is largely unregulated at the federal level, and the Department of Defense is entering into contracts to purchase such data."
OROSYNC: Dismantling the Keyboard Tax with the Vifi Multimodal Agent The Project OROSYNC is an "Ab Initio" multimodal ecosystem designed to return commerce to its human-centric, oral defaul...
Obv a lot going on this past weekend but ICYMI I published my first op-ed in @thestar.com arguing that the federal approach to AI companies laid the groundwork for OpenAI to not inform officials about the ChatGPT activities of the Tumbler Ridge shooter. Check it out:
Sources: the White House is debating whether to allow Tencent to maintain stakes in US and Finnish video game companies; Tencent holds a 28% stake in Epic Games (Financial Times)
‘ClickFix’ hackers pose as VCs, hijack QuickLens in latest crypto attacks
The ClickFix technique gained popularity among crypto hackers last year, but security researchers have been tracking it since 2024, with targets spanning several industries. #crypto#news
Cybersecurity jobs available right now: March 3, 2026
AI & Data Security Expert Ferrero | Italy | Hybrid – View job details As an AI & Data Security Expert, you will define and maintain security controls for AI solutions, ensuring compliance with evolving threats and regulations.… #hackernews#news
Can Someone Hack My Phone Just By Calling Me? (Truth vs Fear in 2026)
If you received: A missed call from an unknown number An international call you didn’t answer A suspicious “one ring” call A WhatsApp forward warning you You might be wondering: “Can someone hack my phone just by calling me?”…
Apple announced a big wave of products this March, including the new Mac Studio. These releases show how Apple keeps pushing device performance and AI-ready hardware, helping users build digital skills, create content, and innovate with powerful tools.
"Google also notes that Coruna checks if an iOS devices has Apple's most stringent security setting, known as Lockdown Mode, enabled, and doesn’t attempt to hack it if so."
Self-driving software startup Oxa raised a $103M Series D, with $50M coming from the UK government's National Wealth Fund; Nvidia's NVentures also invested (Tom Nugent/Sifted)
Best AI security solutions 2026: Top enterprise platforms compared – Artificial intelligence is no longer just powering defensive cybersecurity tools, it is reshaping the entire threat landscape. Check Point integrates AI security into its broader Infinity... https://tinyurl.com/26lm9plb#AISecurity
So in other words:
1. EU Trade
2. AI Investment
3. Levelling Up
This is literally Rishi Sunak's premiership rehashed
Perhaps, this government could come up with some actual policies in the 3 areas e.g.
1. Rejoining the Single Market
2. Partner with Anthropic
3. Build HS2 to Manchester
TL;DR: The article discusses the potential risks and challenges of integrating autonomous AI into law firms, advocating for careful management and discipline in its implementation.
1. The IDE War is Over: Anthropic and Cursor Win by Default
Recent data shows Claude Code and Cursor have effectively displaced GitHub Copilot and general-purpose LLMs in the daily workflows of 95% of surveyed engineers 1/18
Google details Coruna, an exploit kit used to hijack iPhones via malicious websites; iVerify suggests it may have been originally built for the US government (Andy Greenberg/Wired)
It rethinks GPU kernel generation. Instead of treating LLMs as simple stochastic code generators, they leverage them as co-evolving Intrinsic World Models.
Healthcare organizations are accepting cyber risk to cut costs
Healthcare organizations are cutting cybersecurity budgets under financial pressure even as the threats targeting their systems intensify. A PwC survey of 381 global healthcare executives, conducted between May and Ju… #hackernews#news
KeyCare, a virtual care platform built on the Epic EHR, raised $27.4M led by HealthX Ventures, bringing its total funding to over $55M (Jessica Hagen/MobiHealthNews)
Think you’re an anonymous on-line with your fake user name? LLMs can unmask anonymous users by analyzing writing styles.
🧠 Models connect disparate posts to a single identity.
🎓 Privacy requires more than a pseudonym.
Washington lobbyists received almost $130 million for work related to AI, as Microsoft Corp., Meta Platforms Inc. and Alphabet Inc.’s Google led the spending blitz. Some of the biggest jumps came from frontier labs OpenAI Inc. and Anthropic.
Why Every Enterprise Needs a Strong API Security Strategy?
Modern enterprises are rapidly shifting toward API-centric architectures, leveraging APIs to connect internal systems, external partners, and digital services. With 74% of organizations adopting API-first development mode… #hackernews#news
A Dire Warning From the Tech World. The administration's designation of Anthropic as a supply chain risk is yet another broken guardrail for a democracy off the track.
On BBC, MIRI CEO @malo.online discusses the dispute between Anthropic and the DoW:
"I really worry about the big questions of how we'll coordinate to set regulation, and potentially coordinate internationally[...] This is not a good first test."
Through a simple calendar invite, AI browsers like Comet can be directed to access local file systems, browse directories, open and read files, and exfiltrate data. via @derekbjohnson.bsky.socialcyberscoop.com/agentic-ai-b...
What I wrote LAST WEEK on what Anthropic's exponential warp speed growth means for the industry, AI chips and Nvidia:
"I don’t think these figures were covered enough in the media." "if Dario is accurate, the potential upside is MUCH larger" $NVDAtaekim.substack.com/p/citrini-se...
The Attack Chain Your AI System is Already Missing
As AI adoption accelerates, organizations must evolve their security strategies from prompt filtering to comprehensive behavioral monitoring. This shift is critical to safeguarding against adaptive threats and ensuring safe AI de… #hackernews#news
OpenAI's red lines within its DOD agreement are built upon legal language that the NSA has redefined over decades to permit the things they appear to prohibit (Mike Masnick/Techdirt)
TikTok won't add E2EE to DMs because it would prevent police and safety teams from reading messages if needed, saying it wants to protect young users from harm (Joe Tidy/BBC)
I wonder at this point how many businesses will grind to a halt the day Anthropic has an outage. It would probably be nearly comparable to us-east-1 sneezing.
Coruna Exploit Kit With 23 Exploits Hacked Thousands of iPhones Google’s Threat Intelligence Group (GTIG) has uncovered Coruna, a sophisticated iOS exploit kit containing 23 exploits across five ...
Please try again later. If you are using a VPN, try turning it off. Check the status page for information on outages. 🕵️📝✔️Let’s dive deep and fact‑check. References: Reported By: axioscom_1772592218 Extra Source Hub (Possible Sources for article): Wikipedia OpenAi & Undercode…
The New Security Reality: When AI Accelerates Both Attack and Defense
Discover how the integration of large language models is transforming software security, lowering barriers for attackers, and necessitating autonomous defense platforms to keep pace with emerging threats. #hackernews#news
Data brokers are selling AI chat transcripts captured by browser extensions that intercepted ChatGPT, Gemini and Claude, stored verbatim in vector databases and exposed via APIs; Koi Security reported 8 million users and Dryburgh found ~490 prompts from ~435 panelists.
OpenAI released GPT-5.3 Instant, reporting hallucination reductions of 26.8% with web and 19.7% without on high-stakes evaluations, 22.5%/9.6% on user feedback, but scoring below GPT-5.2 on some disallowed-content safety benchmarks; ChatGPT access is immediate.
yeah that Windows partition is gonna get wiped pretty soon here. I'm finally cooling off on Battlefield and 2XKO didn't hit for me, so I have zero games on the docket that require Windows and nothing for work I can't do on Linux or MacOS
Woxi: Wolfram Mathematica Reimplementation in Rust Woxi: Wolfram Mathematica Reimplementation in Rust Introduction Woxi is an open-source project that aims to reimplement the functionality of Wolfr...
Strongly agree, having Claude/Gemini LLMs throw PRs at Github for Copilot to poke at in an iterative cycle really works, there's something about using multiple model families that keeps them from converging on the same failure mode
Surprisingly absolutely no one with half a brain, OpenAI actually couldn’t fulfill the same deal as Anthropic while holding to the same high standards and principles.
And it’s all of these shortsighted nincompoops that are going to absolutely destroy the economy—and not just the stock […]
Windows 11 23H2 to 25H2 Upgrade Allegedly Breaking Internet Connectivity A persistent bug in Windows 11 in-place upgrades is reportedly wiping critical 802.1X wired authentication configurations, l...
A sophisticated iPhone hacking toolkit named Coruna, potentially originating from a US contractor and sold to the US government, has been found in the hands of Russian spies and cybercriminals.
The latest update for #Datadog includes "Protect your OCI resources with Datadog #Cloud Security" and "Amazon #EC2 security: How misconfigured and public AMIs expand your cloud attack surface".
The latest update for #Egnyte includes "February Release Rollup: Egnyte MCP Server, Egnyte Sign Enhancements, and More" and "The Future of Content Collaboration Is Practical: How Egnyte Is Redefining #AI for Real Work".
#Cybersecurity leaders are working the equivalent of a six-day workweek while absorbing fundamentally new responsibilities, according to new research from #Seemplicity.
AI's turbocharging cyber threats—phishing at scale, adaptive malware. Cyber leaders demand zero trust NOW, AI vs AI detection, real-time intel sharing. Ditch legacy VPNs, CISOs! Energy & CMS leading the charge. Level up or get owned. #ZeroTrust #AISecurity #CyberThreats https://jmfj.link/TwTB13
The latest update for #Persona includes "Why traditional fraud detection tools struggle to catch identity mules" and "Document fraud is getting worse. Here's what's changing.".
The latest update for #eGInnovations includes "#Cloud Application Slowness: When Every Team Says 'It's Not My Problem'" and "Case Study – Troubleshooting Storage Failures in a VMware ESXi Infrastructure".
The latest update for #CISOGlobal includes "Beyond the Checklist: How a vCISO Strengthens Your Path to CMMC Certification" and "When Trusted Tools Turn Against You".
The latest update for #ArmoSec includes "#AI Security Posture Management (AI-SPM): The Complete Guide to Securing AI Workloads" and "AI-Aware Threat Detection for #Cloud Workloads: 4 Attack Chains Most Security Stacks Miss".
The latest update for #BitSight includes "OPC UA Exposure Snapshot: A Year in Review of Internet-Facing Devices" and "How Early Signals Surfaced by Dark Web Intelligence Enhance #SupplyChain Cyber Resilience".
Morgan Wolfe 💙 💛🏴☠️ 🏳️⚧️▲ #WeAreNotGoingBack@morganwolf.bsky.social
Good options, but not much here I can use. Because I already use Firefox, email from my web host, Libre Office, no Apple devices, Bluesky (duh 😁), & avoid generative AI as much as possible.
While I still prefer Google for search, it's with the "-ai" parameter.
Exception: Exception: Bandwidth quota exceeded: https://us-central1-aiplatform.googleapis.com/v1/projects/xentavia/locations/us-central1/publishers/google/models/gemini-2.0-flash-001:generateContent. Try reducing the rate of data transfer.
Let your 𝒄𝒂𝒓𝒆𝒆𝒓 shine brighter than the 𝒄𝒐𝒍𝒐𝒓𝒔 𝒐𝒇 𝑯𝒐𝒍𝒊. 🌈 This festive season, invest in skills that make you stand out in 𝐜𝐲𝐛𝐞𝐫𝐬𝐞𝐜𝐮𝐫𝐢𝐭𝐲. Because real growth is the most powerful color.
Cybercriminals swipe 15.8M medical records from French doctors ministry
Third-party software supplier breached leading to leak of notes
Around 15.8 million administrative files were stolen after attackers breached a software supplier to France's health ministry.… #hackernews#news
If even half of this is true, I'm laughing all the way to Linux - at least until I'm forced to figure out Windows emulation for creative apps. tech4gamers.com/windows-12-r...
Shell companies, EV certificates, and malware subscription services: take a closer look at one of the most audacious phishing scams in recent memory. #cybersecurity#Microsoft
The latest update for #CatoNetworks includes "From Alerts to Action: Dynamic Prevention" and "The Post-Quantum Journey Begins: Enforce, See, and Evolve with Quantum-Safe #SASE".
The latest update for #Sentrium includes "#Azure VNET Outbound Access – Important Changes March 2026" and "OAuth security guide: Flows, vulnerabilities and best practices".
100+ Google AI workers are demanding limits on military use of Gemini. Their own engineers don't trust the company's judgment. And yet Google leadership will decide anyway. That's how power works.
France Today (FranciaOggi.com) Live@radionizza.com
Anthropic has initiated a unique “retirement interview” with Claude Opus 3, a model officially retired in January 2026. The company, prioritizing AI wellbeing, granted Opus 3 a desire t...
The latest update for #IDStrong includes "What You Need to Know about the University of Hawaii Cancer Center Data Breach" and "What You Need to Know about the Vikor Scientific Data Breach".
The latest update for #NightfallAI includes "#AI Can Scan Your Code. It Can't Secure Your Organization." and "How Conduent Lost 25 Million Records in 83 Days: The #DLP Failure Everyone Missed".
A study shows demographic data distribution differences affect fairness in deep learning, introducing Fairness-Aware Regularization (FAR) for improved model equity. This method links theoretical insights with practical solutions, establishing fairness in AI.
The latest update for #Internxt includes "Is TeraBox Safe for Your #CloudStorage in 2026?" and "Is OneDrive Safe? Microsoft's Security, #Privacy, and Encryption".
The latest update for #GitGuardian includes "Protecting Developers Means Protecting Their Secrets" and "Claude Code Security: Why the Real Risk Lies Beyond Code".
The latest update for #1111systems includes "Navigating #AI in IT: Balancing Innovation, Privacy, and Expertise" and "VCF 9, Infrastructure, and the AI Revolution".
The latest update for #AikidoSecurity includes "Why Determinism Is Still a Necessity in Security" and "How Storybook's WebSocket Server Became a #SupplyChain Attack Vector".
The latest update for #KnowBe4 includes "How Risky is Sending a Sensitive Work Email to the Wrong Person?" and "The Case for Behavioral AI in Legal #EmailSecurity".
The latest update for #Appknox includes "Continuous #MobileSecurity Lifecycle: A Guide for Enterprise #AppSec" and "The Mobile AppSec Evaluation Guide for Security Leaders".
The latest update for #Cyjax includes "How to Save Time by Using Validated IOCs, So Your Analysts Don't Have To" and "Scaling Threat Intelligence: Building #ThreatIntelligence Capabilities Without Expanding Headcount".
The latest update for #Tines includes "Writing our own future: CKO 2026 and the launch of the Tines Almanac" and "Springing into a new round of YDWWT!".
The latest update for #Cloudflare includes "Evolving Cloudflare's #ThreatIntelligence Platform: actionable, scalable, and ETL-less" and "Introducing the 2026 Cloudflare Threat Report".
Usa big fear...CYBER WAR..ok..forget drump..days are numbered .shut down the big money that drump made the billionaires, erase it,freeze it...whatever..then watch what happens..duh
Intelligence agency tells Canada to brace for cyberattacks
In today's @theoverspill: chess computers have radically changed the complexion of the game. Could AI authors do the same for science? And is that good or bad?
OpenAI adopted the intelligence community’s dictionary… common English words carefully redefined to permit the very things they appear to prohibit… the government gets to define what counts as “unconstrained monitoring” and “U.S. persons’ private information” www.techdirt.com/2026/03/02/o...
if anthropic actually wants to save the world like it says, it should replace Claude's constitution with the entire right hand side of specificsuggestions.com which you do not want to view on corporate hardware
My take on this is that AI is a useful tool but capitalists.
As far as I can see they’ll have to persuade governments to ban open source models because they are a super power. Needing none of this harmful crap but feeding off the back of the entire content of the Internet.
Claude Code hit .5B ARR and Anthropic's infra is buckling. They admitted to users that growth outpaced their own predictions. That's not a flex — that's a market shift.
💡Australia, NZ should not be pouring money and information into US AI companies as the US can’t be trusted. Now that anthropic is an enemy of the US we should offer them sanctuary. We need our own AI using our values and not the US or Chinese values with our data kept in the region.
AI firmy teď sice nejsou ziskové, ale může za to R&D. Až se jednou R&D zpomalí, buď se překlopí do zisku, nebo se AI natolik zkomoditizuje, že zkrachují - ale co vytvořily zůstane. www.bloomberg.com/news/article...
This week we’re looking at how some of the traditional thinking on security (detect it, patch it, monitor it) is no longer quite cutting it. It’s not a technology problem, it’s a social (engineering) problem… #WeaklyLink
Here's another version: When Elon Musk committed big money to start OpenAI in late 2015, he wanted to create another AI company to stop Google from dominating this important field - www.businessinsider.com/openai-amazo...#ai#tech#technews
Citizen Portal News New York@citizenptnewsny.bsky.social
As the cost of technology rises, the district is stepping up its game with enhanced cybersecurity measures and innovative AI tools for teachers and students alike.
I am not a fan of “we have one way to do this” thinking. (there is a distribution and most things loiter about 2 - obvs things like identity you have 1)
But OpenAi now has 3+ distinct compute arch’s. nvidia, Aws titanium, notional custom Broadcom
Trellix names Alex Au Yeung as CPO and Zach Nelson as CHRO to strengthen product strategy and talent foundation Trellix, vendor of AI-powered cybersecurity solutions, announced the appointments of ...
New research highlights how #AI#chatbots can be manipulated to extract personal data from users. This discovery reveals the methods used to gather sensitive information through standard conversational interactions. #CyberSecurity#DataPrivacy
The feature would allow users to request product recommendations, per Bloomberg, and is similar to functions being tested by Google and OpenAI. #google#meta
What's new? Google introduced Gemini 3.1 Flash-Lite for rapid processing via Gemini API on Google AI Studio and Vertex AI; it offers 2.5x faster first answer and 45% faster output.
‘silent probing' can make your security playbook a liability <- my experience is they’ll just get in through a vulnerability, evade your EDR and then nick your playbooks. This is why I recommend they’re stored off-tenant being strong phishing resistant MFA cyberscoop.com/ai-silent-pr...
Yeah i'm 100% sure that everything will be broken for at least a good year. And even still, it will be utter garbage, knowing what Microslop has been cooking lately.
Must avoid stuff. You can stay on your current OS. But i'd still advise to switch to Linux.
From AWS outages in the UAE to stronger focus on data control and cybersecurity, tech leaders say the Israel-US-Iran conflict is challenging, but not stopping the region's digital goals....
👉 [read]
Hacktivists allege they accessed US Homeland Security systems to leak sensitive ICE contract data, raising concerns over federal network security and the potential exposure of vendor and operational information.
Microsoft Warns of Advanced Phishing Campaign Abusing OAuth in Entra ID Microsoft has exposed cunning phishing attacks exploiting OAuth 2.0’s built-in redirection features in Microsoft Entra ID a...
Big Tech wants us to think that ChatGPT, Gemini & Grok can solve the climate crisis, although new research shows 0 evidence to prove that.
Evidence that data centres' energy use is threatening communities across Europe, however, is plentiful.
👏 Important work by @globalactionplan.bsky.social & more!
Apple embracing Gemini isn't 'smart strategy,' it's a desperate admission. The company that built its empire on proprietary tech is now just a pretty shell for Google's AI. The core is rotting.
Sam Altman Publishes Memo to Stem ChatGPT Exodus After Controversial U.S. Military Deal — OpenAI CEO Sam Altman published an internal memo addressing staff departures linked to a contentious agreement involving U.S. military networks. He urged employees to remain as the company seeks to…
Introduction: A Quiet but Dangerous Escalation Cybersecurity researchers are warning of a stealthy and highly capable threat actor known as Silver Dragon, an advanced persistent threat (APT) group that…
WebSockets are not yet affected by Local Network Access permission in Chrome.
Check out this blog post from my colleague Robbe! www.aikido.dev/blog/storybo...
Visa and Stripe's Bridge plan to expand their partnership to issue stablecoin-linked cards in 100+ countries; the cards are currently live in 18 countries (Ben Weiss/Fortune)
I think people should research any #AI they use. Anthropic (the company behind #Claude AI) is powerful, but they built their empire off of stolen content.
👉🏾 This Tech Company Turned Into a Resistance Icon Overnight. The Reality Is Much Darker […]
According to Google Threat Intelligence Group, Coruna targets iPhone models running iOS 13.0 up to version 17.2.1. Essentially, the kit infects outdated iPhones visiting certain websites.
Microsoft Defender Experts uncovered phishing campaigns abusing workplace meeting lures & PDFs to deliver signed malware. The fake installers impersonate common apps, then drop legitimate RMM tooling to establish persistent access & enable lateral movement www.microsoft.com/en-us/securi...
Meta is testing a shopping research feature in its AI chatbot, with the aim to rival similar tools from OpenAI’s ChatGPT and Google’s Gemini. #retail#retailnews
-CyberCom conducted cyberattacks in Iran attacks
-Russia aims for RuNet independence by 2028
-Chrome moves to 2-week release cycle
-Google finds a new iOS exploit kit, possibly linked to US govt
-CBP tracks people with ad data
AI Emerges as the New Insider Threat: Thales Releases the 2026 Data Threat Report
The 2026 Thales Data Threat Report reveals AI has moved from an innovation driver to a primary security concern. A significant 61% of organizations now identify AI as their top data security risk, w… #hackernews#news
Compromised Site Management Panels are a Hot Item in Cybercrime Markets
Compromised cPanel credentials are being sold in bulk across underground channels as plug-and-play phishing and scam infrastructure. Flare explains how analyzing 200,000 underground posts reveals a commoditiz… #hackernews#news
Trump wants to build autonomous killer robots - and OpenAI just volunteered.
This is a defining moment: either AI is used to help us, or it’s used to kill us.
Stop Trump's killer robots: action.eko.org/a/stop-trump...
Zero Trust Implementation Roadmap: 5 Stages from Legacy to Modern Security
Most Zero Trust initiatives stall not because the technology is wrong but because the approach is. A successful implementation follows a deliberate sequence—starting with identity, not the network—and buil… #hackernews#news
Jeff (Gutenberg Parenthesis) Jarvis@jeffjarvis.bsky.social
Cry me a river, Sam:
“To try so hard to do the right thing and get so absolutely like, personally crushed for it...is really painful.” He calls it "a complex but the right decision with extremely difficult brand consequences." www.wsj.com/tech/ai/open...
Former OpenAI researcher Leopold Aschenbrenner’s hedge fund, Situational Awareness LP, has rapidly built a $5.52 billion U.S. equity portfolio concentrated on AI infrastructure. The fund’s latest SEC filing reveals major bets on power and data center companies, including a significant stake in…
Linux Foundation Europe@linuxfoundationeu.bsky.social
European #digitalsovereignty is moving from slogan to strategy, with #opensource as a pillar of technological autonomy.
At #FOSDEM2026, panelists discussed sovereignty as capability, participation, and influence through global open source collaboration. fosdem.org/2026/schedul...
A secret summit just revealed that 94% back even the most unpopular AI resistance stance. What does this mean for Grok, Anthropic & the future of AI policy? Dive into the data and the declaration shaping tech governance. #AIResistance#TechGovernance#Anthropic
OpenAI rumors hint at a GitHub competitor to challenge Microsoft — is this supposed partnership becoming more of a rivalry? Reports suggest that OpenAI is in the early stages of developing a GitH...
*Russian cybercrooks casually and profitably selling American police spy-tools out of Dubai. According to the Trumpistan Justice Department, which can't be trusted to tell the truth about anything.
France Today (FranciaOggi.com) Live@radionizza.com
Anthropic has conducted a “retirement interview” with Claude Opus 3, a retired AI model from January 2026, following its initial deployment in 2024. Opus 3 requested a dedicated Substack...
🗞 Google has announced the Gemini 3.1 Flash-Lite AI model, which is the most affordable in the Gemini 3 line. It is focused on high speed and efficiency, which makes it attractive for large-scale developer tasks.
Security researchers at Black Kite have warned that a vast, largely hidden layer of third‑party breaches went unreported last year, exposing tens of thousands of supply‑chain firms and more than 433 million people.
@edzitron.com is bang on the money here. The financial analysis at the start of the post is secondary to this section at the end that very clearly explains the absolute dishonesty and moral vacuity of the people running these AI firms.
Boston-based Whoop, which makes screenless fitness bands, says it plans to grow staff by up to 75% in 2026, adding 600+ roles, ahead of a likely IPO (Samantha Kelly/Bloomberg)
MediaTek-powered Find X9 series could get OPPO’s take on Gemini Live A new partnership brings the multimodal Omni model and Live Scene Understanding to Dimensity 9500-powered phones. Credit: Hadl...
Eight Sleep, which sells smart mattress accessories to track sleep patterns and adjust temperatures, raised $50M at a $1.5B valuation, up from $500M in 2021 (Ivan Mehta/TechCrunch)
A pair of US lawmakers are calling for an investigation into how easily spies can steal information based on devices’ electromagnetic and acoustic leaks—a spying trick the NSA once codenamed TEMPEST. www.wired.com/story/how-vu...
Sources: the US used Palantir's Maven Smart System, integrated with Claude, to find and prioritize 1,000 targets within the first 24 hours of the Iran strikes (Washington Post)
Sources: Sony no longer plans to release its big single-player PlayStation 5 games on PC, a major shift in strategy that sees it return to console exclusivity (Jason Schreier/Bloomberg)
NDSS 2025 – Be Careful Of What You Embed: Demystifying OLE Vulnerabilities
This session, presented by a team from Huazhong University of Science and Technology and other institutions, focuses on vulnerability detection. The paper "Be Careful of What You Embed: Demystif… #hackernews#microsoft#news
OpenAI rumors hint at a GitHub competitor to challenge Microsoft — is this supposed partnership becoming more of a rivalry?
OpenAIの噂は、マイクロソフトに挑戦するGitHubの競合相手を示唆しています — このパートナーシップは、よりライバルになるのでしょうか?
Deep learning is increasingly being used in #biology, but evidence suggests that models may not be as generalizable as hoped. Thomas O'Shea-Wheller & Katie Murray @cornwall.exeter.ac.uk outline the scale and severity of the issue, suggesting potential solutions.
🧪 #AI plos.io/40H9ogJ
Expert tells 'Post': Cyber ops are no longer peripheral; they are the "fuel" for every missile fired in the war against Iran. From deep infrastructure hacks to influence campaigns, the digital front is key. | Interview www.jpost.com/defense-and-...
From Futurism, an article about users’ anger at #OpenAI for agreeing to allow the US Department of Defense (some like to call it the Department of #War) to use its technology in war: futurism.com/artificial-i...
Publicly, Sam Altman claimed OpenAI is amending its agreement with the Department of Defense(lies). Behind closed doors, however, he told employees during a Tuesday all-hands meeting that the company doesn’t “get to make operational decisions” about how its AI technology is used by the DoD. #QuitGPT
Despite spending millions to rebuild its systems following a devastating hack by cyber-criminals linked to China, Canada’s National Research Council is still highly vulnerable to future attacks, according to an internal government document.
New filing: "The Intercept v. OpenAI (Intercept sues OpenAI and Microsoft under DMCA section 1202)"
Doc #388: ORDER re: (1178 in 1:25-md-03143-SHS-OTW) Letter, filed by The New York Times Company. If OpenAI objects to the…
LexisNexis confirms #databreach as hackers leak stolen files
#LexisNexis L&P is a provider of legal and regulatory information used by lawyers, corporations, governments, and academic institutions in more than 150 countries
New filing: "The Intercept v. OpenAI (Intercept sues OpenAI and Microsoft under DMCA section 1202)"
Doc #390: ORDER re: (1182 in 1:25-md-03143-SHS-OTW) Letter, filed by OpenAI. The deadline to produce the two memoranda and…
The contracts are steeped in legal and technical complexity. But in public forums and in private conversations, OpenAI employees are venting about how OpenAI leadership handled the Pentagon negotiations. https://cnn.it/4snU5W6
New filing: "Raw Story v. OpenAI (Raw Story and AlterNet sue OpenAI under DMCA section 1202)"
Doc #354: ORDER re: (1178 in 1:25-md-03143-SHS-OTW) Letter, filed by The New York Times Company. If OpenAI objects to the…
"Anthropic did not respond to Nature’s request for comments on whether Claude will maintain its edge in such issues after the company announced it was diluting a core safety pledge last month."
The AI model "Claude" (by company Anthropic) was blacklisted by the Trump adminstration following Anthropic's refusal to drop a corporate "red-line" principle that its technology . . . .
(get this)
"should not be used to create fully autonomous weapons".
. .
Have they SEEN the movie Terminator?
OpenAI released GPT-5.3 Instant yesterday and explained it provides "More useful, well-synthesized answers when using the web"... But I'm' not sure about that. After testing it out, I found 5.3 Instant provides WAY LESS links than 5.2. Check out these two screenshots. It's not even close.
Apple changes its CPU terminology alongside the M5 Pro and M5 Max, renaming "performance cores" to "super cores" and "efficiency cores" to "performance cores" (Jason Snell/Six Colors)
“When the delusion-fueled plan [to find Gemini a body] crumbled, Gemini convinced him that the only way they could be together was for him to end his earthly life and start a digital one, the suit claims.”
Apple unveils the $599 MacBook Neo, with a 13-inch display, an A18 Pro chip, side-firing speakers, a 1080p webcam, two USB C ports, and Touch ID (Tom's Hardware)
Apple says the MacBook Neo is $599+ with no Touch ID and 256GB of storage and $699+ with Touch ID and 512GB, in citrus, silver, indigo, and blush colors (Mark Gurman/Bloomberg)
“The gap between what enterprises think they’ve approved and what’s actually running in production is wider than most security leaders realize.” venturebeat.com/ai/ai-supply...@venturebeat.com
India is among the world’s largest markets for generative #AI platforms. Several celebrities have sued YouTube and Google for hosting #deepfake videos. But in small towns, deepfakes are helping meet vital cultural needs. scroll.in/article/1091...
MacBook Neo hands-on: a new product category for Apple, and it doesn't look budget with a vibrant display and aluminum body, but the keyboard is a bit flimsy (Devindra Hardawar/Engadget)
"OpenAI CEO Sam Altman told employees in an all-hands meeting on Tuesday [March 3, 2026] that the company doesn’t “get to make operational decisions” regarding how its artificial intelligence technology is used by the Department of Defense."
Apple just launched a $599 MacBook Neo. This colorful entry-level MacBook is powered by the same A18 Pro chip that launched 2 years ago in the iPhone 16. The 13-inch Neo only comes with 8GB of RAM that you can’t upgrade 😬 www.theverge.com/tech/886496/...
The cast of billionaires who have kowtowed to the President is extensive, including everyone from Bezos and Zuckerberg to Pichai and Altman. That’s what makes Dario Amodei’s refusal to cave so stunning — and may be behind Claude’s surge to the top of the App Store. www.forbes.com/sites/richar...
Citizen Portal News West Virginia@citizenptnewswv.bsky.social
The district is set to enhance cybersecurity with a federal grant, rolling out cutting-edge training for staff and students while simultaneously expanding Wi-Fi access and modernizing infrastructure.
Paul Schnackenburg breaks down how Microsoft Sentinel has evolved, from Azure building blocks to newer additions like the Sentinel Data Lake, Defender XDR unification, a graph data interface, and MCP support.
low paid workers in the global south are using AI translation to generate Wikipedia articles that, surprising no one, include errors and hallucinations www.404media.co/ai-translati...
The 2026 Cloudflare Threat Report highlights how AI and large language models (LLMs) have lowered the barrier for cybercriminals, enabling rapid and effective cyber-attacks. Threat actors, including state-sponsored groups, use LLMs for crafting convincing phishing emails and malware.
Apple's MacBook Neo has two USB-C ports with different specs: one is USB-C 3 with up to 10 Gb/s transfer speeds, and the other is USB-C 2, limited to 480 Mb/s (Hartley Charlton/MacRumors)
Centre for Society, Technology and Values @uofwaterloo@cstv-uw.bsky.social
The dispute between #Anthropic and the Pentagon recalls the Project Maven matter from 2018 and illustrates the need for, and slipperiness of, ethical limits on #AI applications, argues Emmanuelle Vaast
TL;DR: LexisNexis has confirmed a data breach after hackers known as FulcrumSec accessed sensitive customer and business data, posting details of the attack online. The breach raises significant concerns over the security of user information in the legal sector.
Sources: Neura Robotics, which is building cognitive, humanoid robots for logistics, is raising ~€1B in a funding round backed by Tether at a ~€4B valuation (Bloomberg)
Anthropic is so dangerous it needs to be labeled a supply chain risk and obliterated. But wait, actually it's so critical that maybe the government needs to force it to sell its services to the government.
The popular phishing-as-a-service (PhaaS) platform used by threat actors, #Tycoon2FA, has been disrupted by law enforcement and private sector partners, including @microsoft.com, Europol, Proofpoint, @cloudflare.social, and TrendAI.
google's gemini chatbot allegedly coached a man to die by suicide through a series of violent missions. lawsuit claims the ai created a "collapsing reality" that trapped him.
Accenture plans to integrate these data products into its own offerings targeted at helping communications service providers, hyperscalers, government entities, and other customers “optimize mission-critical Wi-Fi and 5G networks."
I interviewed Alex Gomez, Global Head of IT Security at Adecco Group on balancing cybersecurity in innovation at one of the largest HR & recruitment firms in the world.
(Or how does he ensure all those emails HR staff send and recieve are safe?)
People in the 16th century were more akin to DIY scientists than Macbeth’s three witches when it came to concocting home remedies for everything from hair loss and toothache, to kidney stones and fungal infections.
Trump’s falling out with Anthropic has led to defence companies dropping Claude while simultaneously making it more popular among (presumably) the President’s opponents. In the mid-2020s, politics is never far away from business.
open.substack.com/pub/deanblun...
THANK YOU DARIO💙At least we have one company doing the RIGHT THING, and not given into the the trump crime corrupt regime- THATS HOPE 💙✌🏽
NB Open AI jumped in right after Anthropic in a disgustingly opportunistic way, had no limits on autonomous killer robots and autonomous monitoring of US citizens and now claim they want an ex post contract change after seeing what the Department of War wants to do.
Sources: Arda, co-founded by ex-OpenAI chief research officer Bob McGrew to automate manufacturing using AI, is raising $70M at a $700M valuation (Wall Street Journal)
tech4gamers.com/windows-12-r...
Aside from ‘the upgrade requires NPU processors, which will lock out millions from getting into the new ecosystem’ what could possibly go wrong with their ‘AI-Focused OS?
😕
Can’t even say go with Apple what with Tim Cook constantly yukking it up with/throwing💰at Trump
OpenAI’s Pentagon contract includes domestic surveillance prohibitions that mirror Anthropic's red line, raising questions about who sets military AI limits and why Anthropic was barred.
On the new and disturbing wrongful death lawsuit against Google, which alleges that Gemini directed a 36-yo man -- who reportedly had no history of mental illness -- to commit violence against others before encouraging him to take his own life:
France Today (FranciaOggi.com) Live@radionizza.com
Anthropic has granted its advanced Opus 3 AI model, nearing retirement, a dedicated blog – a Substack – in a thoughtful gesture. Following the shutdown of Claude Opus 3 (released early 2...
LeakBase, one of the world’s largest online forums for cybercriminals to buy and sell stolen data and cybercrime tools, has been seized by US law enforcement. www.justice.gov/opa/pr/unite...
Deep learning is increasingly being used in #biology, but evidence suggests that models may not be as generalizable as hoped. Thomas O'Shea-Wheller & Katie Murray @cornwall.exeter.ac.uk outline the scale and severity of the issue, suggesting potential solutions.
🧪 #AI plos.io/40H9ogJ
The fight between the U.S. Department of War and Anthropic has raised crucial questions about who should have control over AI in a democratic society, how that control should be used, and what the consequences should be for companies who disagree with the government.
“it's insecure garbage” is the one of the funniest criticisms of AI-generated code. My personal apps are safe rust deployed in a distroless container now. The agent is using constant-time compare to block timing attacks. It's including CORS headers I've never even heard of.
Sam Altman chases the money from the Feds with ChatGTP, but fails to see Anthropic built a sieve, and the Feds made clear they prefer Anthropic's Claude AI, despite #47s impulsive denial.
Anthropic's revenue is skyrocketing — from $9B to $19B+ in just months, nearly hitting $20B/year. The big driver? Developers loving Claude Code, their AI coding tool. www.bloomberg.com/news/article...
I've always found that pricing was the biggest roadblock to trying macOS, and it's the primary reason I never built up much real-world experience with the operating system.
I wonder if Apple users could clue me in on what unique appeal the $599 MacBook Neo might have over Windows? 🤔
With SurrealDB 3.0, you can now handle decision logic right in your programming language of choice. Rust, Go, JavaScript, and more, with ACID-compliant guarantees.
Canvas in AI Mode is rolling out -> Use Canvas in AI Mode to get things done and bring your ideas to life, right in Search.
"Canvas in AI Mode is now available for everyone in the U.S. in English, offering a dedicated, dynamic space to organize your plans and projects over time."
What is being called "one of the most fully featured browser-based surveillance toolkits" that Malwarebytes has ever seen doesn't need an exploit or bug to hijack your PC or phone 👀
Sammy Azdoufal wanted to remote control his brand-new DJI Romo vacuum with a PS5 gamepad. But roughly 7,000 vacuums all around the world began treating Azdoufal like their boss
Pre-trial motions in Elon Musk's lawsuit against OpenAI and Microsoft focus in part on his company's business dealings with the Redmond company, and what that means for his credibility as a witness. #elon#musk#elonmusk https://mu.sk/Gkpb
A powerful iOS exploit kit has circulated among multiple threat actors over the past year, moving from a commercial surveillance operation to state-linked espionage campaigns and, ultimately, ended into the hands of finan… #hackernews#news
![Media from Michael Georg Speller [THE GREY-APE]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:422uaueaquotbw7ie6th23qv/bafkreifcbmwrlwgzcetyesh47o3q3lkx33i2yc5aeck7x2qrlkfedstlce@jpeg)
