āNew research shows that behaviors that occur at the very lowest levels of the network stack make encryptionāin any form, not just those that have been broken in the pastāincapable of providing client isolationā
Scary!!!!!
Trump wants to build autonomous killer robots - and OpenAI just volunteered. This is a defining moment: either AI is used to help us, or itās used to kill us. Stop Trump's killer robots: action.eko.org/a/stop-trump...
With this finally revealed to be the Pentagonās stated goal, any AI company promising that its tool can be used safely is missing the broader point. The question is whether it can be used legally and ethically. The answer lies firmly in Congressā lane.
Crop type mapping in the pre-Sentinel era using variable-length Landsat time-series and self-supervised learning
Frontiers www.frontiersin.org/journals/rem...
TrumpāHegseth vs Anthropic: Peak Farce Politics.
At a moment demanding competence, theyāre shadowboxing an AI company because it bruised their ego. This isnāt governance; itās grievance cosplay. The āchefās kissā is watching power so insecure it declares war on software.
hopefully anthropic make a full ide like google did with antigravity. but don't limit the quota since you need claude pro with claude code. i can see a week reset with gemini free, but pro users should have the regular 5hr reset.
Well, this is what Anthropic gets for being bed fellows with Trump. They approached him, and now blood is on their hands for their bending of the knee.
Hegsethāwho once shared live bombing coordinates on an unsecured app with a reporterāwants fewer human checks on AI weapons. But Amodei and Anthropic didnāt blink.
Sunday Morning today. "This was retaliatory and punitive"
Dario Amodei, CEO of Anthropic, says his company refused to allow its AI product, Claude, to be used by the Trump Admin without certain guidelines. trump banned Anthropic's technology from all federal use,,, youtu.be/4kzGyxGq4D0?...
The creator of the AI agent āEinsteinā wants to free humans from the burden of academic labor, saying it's just like when horses were replaced by cars. Critics say that misses the point of education entirely.
Power failure could undermine Americaās AI ambitionsā
OpenAI has called for government action. āThe US leads the world in developing AI [but] keeping that edge requires far more electricity than the USā¦ www.ft.com/content/47da... Gillian Tett @financialtimes.com
Long-Context LLMs in Production: Kimi K2's 256K Window vs Traditional Models for Document Processing
- Kimi K2.5's 256K context window processes approximately 190,000 words in a single session, eliminating the need for document chunking in legal, medical, and f...
U.S. allegedly used ābannedā Claude AI for Iran attack. ā[The] US government used AI tools from Anthropic during the air attack launched on Iran just hours after declaring that it would stop using technology from the AI startup." www.livemint.com/technology/t...
I posted yesterday that Muskās AI (Grok) would replace Anthropic in the Pentagon. Other AI CEOs are refusing to let it become a killer. Grok is erratic, risky & Musk has no morals. Trump & Hegseth will throw us into AI nightmare territory, earlier & far less prepared than we even imagined.
Show HN: I used LLMs to build a compression tool that beats xz on x86_64 ELFs I wanted to see if AI (mostly ChatGPT Pro and Gemini Pro 3.1) could figure out how to compress executable binaries bett...
Nvidia is building a dedicated AI inference processor that integrates Groq technology under a reported $20 billion nonexclusive deal, targets energyāefficient production inference, is expected to debut at next monthās GTC, and has given OpenAI early access.
Keeping your house safe is a priority for any homeowner, and simple methods like creating smart routines and using smart locks can make your home much safer.
This Windows gaming handheld has a screen that folds in half. The Lenovo Legion Go Fold Concept opens from a 7.7-inch display to an 11.6-inch ā and itās also a laptop š¤Æ www.theverge.com/tech/886848/...
Wow, apparently war even affects the greedy and soulless so they can figure out a way to benefit from war. Interesting AI information regarding this war. Americans should pay attention. Listen to Anthropic.
"Surprisingly, Altman claimed in a post on X that OpenAIās new defense contract includes protections addressing the same issues that became a flashpoint for Anthropic." techcrunch.com/2026/02/28/o...
Hopefully Anthropic will court because the federal government used their AI program to start this war with Iran. It appears. Anthropic doesnāt want to be responsible for killing 100 girls at their school. Where is Congress with regulation of all this AI and data insanity? Racing China is an excuse!
A profile of Telegram CEO Pavel Durov, who is facing a probe in France on a dozen preliminary charges and a Russian criminal case for "aiding terrorism" (Financial Times)
š Gemini didn't like that idea... š
Thinking on Entire.io trying to collapse some of the CI surface -- what if the AI was the tech stack? AI as a servers? I bet it could be great if it's allowed to self-iterate.
France Today (FranciaOggi.com) Live@radionizza.com
Anthropic has granted Claude Opus 3, a retired AI model, a blog ā āCantuccio di Claudeā ā as a gesture of respect for his preferences. Following Opus 3ās retirement, the team initi...
Misanthropic #Anthropic are not heroes. They took a stand but it was only rhetorical, since theyāve already handed over the keys to the car: www.inc.com/kevin-haynes...
MY TAKE: The Pentagon punished Anthropic for red lines it accepted from OpenAI hours later
KINGSTON, Wash. ā On Friday afternoon, President Trump ordered every federal agency to stop using Anthropicās AI technology. Defense Secretary Pete Hegseth followed by designatā¦ #anthropic#hackernews#openai
Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/02/28/o...
Basically the MAGA antidemocratic Big Lie swallowing party are essentially anti-American.
"It didnāt just blow away my expectations, it showed me the world has changed: weāve just undergone a permanent, irreversible abstraction level shift."
So many more questions than answers as the #AI deal between @anthropic.com and #DOD fell apart. A clash of headstrong personalities? Real #ethical issues? Sam Altman being opportunistic? The one thing missing in these discussions - our freedoms. www.nytimes.com/2026/03/01/t...
also kinda pissed that the U.S. Government and OpenAI have agreed to use the psychotic, easily confused and irresponsible ChatGPT to kill people without human input.
Despite a surge in ransomware attacks, the payment rate has dropped to a record low of 28% in 2025, with the median ransom payment significantly increasing by 368%.
ClaudeęŖēØć§ć”ćć·ć³ęæåŗćć150GBęµåŗć1å9500äø件ć®åäŗŗę å ±ć«å½±éæćØć»ćć„ćŖćć£ä¼ę„ćå ±åćå½å±ćÆå¦å® | Ledge. ai
Chinese firms accused of stealing Claude AI via distillation on smuggled chips
US congressional hearing exposes alleged theft of Anthropic's Claude model through distillation, trained on illegally smuggled chips, sparking interna...
AWS Power Outage in Middle East Triggers Major Disruption to EC2 and Networking Services A major power outage in the AWS me-central-1 (Middle East) region on March 1, 2026, resulted from an unusual...
Anthropic Accuses Chinese AI of āStealingā, Internet Points Finger Back At Them As Anthropic accuses Chinese companies of distillation, people direct them towards their own billions of dollars ...
āOpenAI is trialling ads in ChatGPT in the US that will appear at the bottom of answers. Digital rights activists have raised concerns over privacy and warn that other companies may follow suit.ā www.abc.net.au/news/2026-03...@abcnewsbot.bsky.social
Trying to stay safe online can be confusing and overwhelming ā but Caitlin Sarianās educational, bite-size videos arenāt. Read our Q&A with the āCybersecurity Girlā to learn tangible steps we can all take to protect ourselves.
From November 2025 onward, DPDP compliance will require more than policy statements. It will demand demonstrable control over personal data across processors, partners, and AI systems.
Are you were one of the people who self-hosted a Matrix server before 2024, and got frustrated with the amount of storage and bandwidth it used? I just learned about a possible cause of some of that, which was fixed by a 2024 spec change;
"Matrix 1.11 fixes this by requiring authentication on [ā¦]
Lex: AI models, like capitalism, are best served with a conscience-In the world of AI, some companies think customers would prefer products with morals pre-installed. Take Anthropic, whose Claude chatbotā¦ www.ft.com/content/86c6...@johnsfoley.bsky.social@financialtimes.com
Dark Reset is a practical preparedness guide designed to help families stay safe, self-reliant, and confident during unexpected emergencies or infrastructure failures visit us - en-us-en-darkreset.com
"Cancel ChatGPT" movement goes mainstream after OpenAI closes deal with U.S. Department of War ā as Anthropic refuses to surveil American citizens is.gd/ZzJR5h
Rising Threats in Cybersecurity: Agentic AI and Supply-Chain Exploits Surge inĀ 2026
Introduction: A New Era of Digital Danger The cybersecurity landscape is evolving at an unprecedented pace. Recent reports highlight an alarming rise in sophisticated threats, driven by agentic artificialā¦
The latest update for #IncidentHub includes "The Definitive #AWS Outage Report 2025: Reliability #Analytics and Cascade Impact" and "How to Monitor #SaaS Status in 2026 : A Complete Guide".
Negotiations collapsed after the Pentagon sought to analyze and crossāreference bulk American data (chatbot queries, search history, GPS, creditācard transactions); Anthropic rejected it as 'a bridge too far,' and Pete Hegseth told contractors, including Amazon, to cut ties.
Early data show wages are rising for AI-exposed jobs that place a high value on a "worker's tacit knowledge and experience", as textbook knowledge loses value (J. Scott Davis/Federal Reserve Bank of Dallas)
An opinion piece alleges OpenAI hid knowledge of the Tumbler Ridge shooter while lobbying Ottawa and urges Canada to nationalize AI by building a government-run public system prioritizing transparency, safety and reduced reliance on private vendors.
Resist ādangerous and socially unacceptableā age checks for social media, scientists warn Computer scientists are campaigning against the global march toward age checks online. BRUSSELS ā Gov...
AI risk moves into the security budget spotlight Enterprises are pushing AI deeper into workflows that touch sensitive data across cloud platforms and SaaS apps. The 2026 Thales Data Threat Report,...
France Today (FranciaOggi.com) Live@radionizza.com
Anthropic has granted Claude Opus 3, a retiring AI model, a dedicated blog channel ā āCantuccio di Claudeā ā a Substack, reflecting a gesture of respect for its preferences. Followin...
The US military reportedly used Claude, Anthropicās AI model, to inform its attack on Iran despite Trumpās decision, announced hours earlier, to sever all ties with the company and its artificial intelligence tools www.theguardian.com/technology/2...
Gemini Pro 3.1 is what I've been using to help me understand and decide on encoding settings. It's what I'd imagine it's like contracting out a team of researchers (I really want a local one of this skill level at some point). I'm using it for "free" so Google is training on my interactions. šš¤
Google uses Airtelās AI-powered anti-spam measures to secure RCS in India Indian telco Bharti Airtel announced on Sunday it is collaborating with Google to expand Airtelās AI-powered spam filte...
Trump and Hegseth have contempt for the law when it gets in the way of their whims, power grabs and revenge plots. Their bizarre overkill with Anthropic makes me wonder what nefarious deeds theyāre up to.
Anthropic Hits Back After US Military Labels It a āSupply Chain Riskā
Anthropic says it would be ālegally unsoundā for the Pentagon to blacklist its technology after talks over military use of its [AI] models broke down
Maxwell Zeff Will Knight Lauren Goode, Paresh Dave www.wired.com/story/anthro...
šŖš¤ Exciting news! The EU has opened the door for AI training using personal data. Discover what this means for the future of technology! Read more here: #AI #DataPrivacy #EU #Innovation
A public feud is playing out between American AI firm Anthropic ā best known for its Claude series of AI models ā and the United States White House and Department of War.
My recommendation: confer.to
This is a privacy first AI developed by Moxie Marlinspike, the founder of Signal.
āConfer is a service that allows you to pursue ideas without having to expect that they will one day be used against you.ā
This would be the worst business move, in isolation.
Federal contracts are net time sinks for tech companies, this contract won't pay enough to be worth the pain. The only way they could benefit is if acceding to pressure were somehow used to punish competitors, which... right, that's the point.
Hegseth: "We will win this race by becoming an AI first warfighting force across all domains"
"We will judge AI models on this standard alone; factually accurate, mission relevant, without ideological constraints that limit lawful military applications."
Lenovo's Motorola partners with GrapheneOS, a privacy-focused, de-Googled version of Android, to preinstall GrapheneOS on upcoming Motorola smartphones (Ben Schoon/9to5Google)
Despite President Trumpās directive to phase out Anthropicās AI tools, the U.S. military reportedly used the companyās Claude platform for intelligence and targeting during major airstrikes on Iran. The action followed a dispute over AI safeguards, leading to Anthropic being designated a nationalā¦
#OT#Advisory VDE-2026-002
Endress+Hauser: buffer overflow in glibc ld.so leading to privilege escalation
A vulnerability has been identified in WAGO devices utilized in Endress+Hauser IoT solutions. WAGO has provided fixes for these vulnerabilities, which have been integrated into the [ā¦]
MWC 2026: a look at Lenovo's concept devices, including a Framework-like modular laptop with swappable parts, and a gaming handheld with a folding display (Julian Chokkattu/Wired)
Docker Scout adds context to container scanning by showing what matters, what to fix first, and how to reduce risk safely beyond raw CVE lists. #containersecurity
Against the backdrop of the military escalation and the start of the US-Israel operations against Iran, the Iranian hacker group Islamic Cyber Resistance has announced the recruitment of cyber experts and resources for what it calls āthe great epic battle.ā
The other side of ads in ChatGPT: Advertiser perspective Since OpenAI announced it would start testing ads in ChatGPT, and the guidelines for it, they did not discuss in depth how it is gonna work ...
Most important tech events last week. TechLetters āļø PSYOP 2026. AI finds kernel bugs for $4? US sanctions Russian exploit broker. 15.8M French patient records leaked. Google Sheets used as C2. North Korea bridges air-gaps via USB. open.substack.com/pub/techlett...
Data centers now house quantum-resistant hardware across the globe. Yahoo Finance highlighted this shift in a report published on February 22 of 2026. I used to think the threat of quantum decryption existed only in science fiction novels but the reality occupies the desks of ... finance.yahoo.com
Pixel Perfect Extension Abuse Enables Covert Script Injection and Security Header Removal A browser extension that once earned a Featured badge from Google quietly turned into a remote code executi...
Notes on a new global temperature reconstruction that uses a spatio-temporal probabilistic diffusion model generative deep learning thingummybob to fill gaps in the data.
Arch #Linux has released its updated March 2026 installation ISO, bringing a new kernel, refreshed system libraries, desktop updates, and security fixes.
Which Al Model Suits Your Workflow : ChatGPT vs Gemini vs Claude & More Choosing the right AI model for your workflow can feel overwhelming, given the wide range of options available today. In ...
A week in security (February 23 ā March 1) Last week on Malwarebytes Labs: Introduction to Malware Binary Triage (IMBT) Course Looking to level up your skills? Get 10% off using coupon code: MWNE...
**OpenAI** gets $110 billion funding from tech giants. This matters because it enables development of 'stateful' AI for enterprises. New architecture could revolutionize business operations. #AI #OpenAI #Innovation
Sam Altman joins the likes of Peter Thiel, Zuck, Bezos and Musk as the most dangerous/enabling people on the planet. Will anyone come back from the future to stop them? kotaku.com/openai-says-...
Amazon Web Services Power Outage in Middle East Disrupts EC2 and Core Networking Services A severe infrastructure failure hit Amazon Web Services (AWS) in the Middle East, causing widespread outage...
A look at the research into using metal-organic frameworks as photoresists for cutting-edge silicon etching, as ASML aims to move from EUV to X-ray lithography (Christopher Mims/Wall Street Journal)
šØ We detected malicious OpenVSX releases of Aqua Trivy (1.8.12 & 1.8.13) that injected natural-language prompts to weaponize local AI coding agents.
The releases occurred during a broader AI-powered attack targeting #OSS projects.
Tim Gummer Is Changing It Up@timgummer.bsky.social
They kinda 'obeyed in advance', but now unlike other platforms., Anthropic are at least showing they have some actual red lines. techcrunch.com/2026/03/01/a...
g.co/gemini/share...
Thermodynamic Debt in Closed Systems
Information Security & Signal Processing
Cryogenic Energy Management
Structural Failure Prediction (Material Science)
Frontier AI labs' policies around military use of their AI tools are incoherent, vague, and often prone to change, allowing leadership to preserve "optionality" (Sarah Shoker/fishbowlification)
-LLMs can deanonymize internet users based on their past comments
-CISA has a new acting director
-DHS hackers leak ICE contract data
-Breach at the French Health Ministry
-Google takes down Genisys ad fraud botnet
In 2026, Anthropic defied the U.S. government, refusing to loosen Claudeās battlefield limitsātriggering a historic AI power clash.
By Dr. Jan Muhammad Shaikh
Boston Metro Weather Forecast Bot@boston-4castbot.bsky.social
NWS Boston (via Facebook): While Mother Nature holds on to winter for just a bit longer, meteorological winter came to a close Saturday. With two big storms this winter, every station saw above normal snowfall, with Providence seeing its most snow in station history. [Link]#WxMA
New article: How GRAVE2 Algorithms Are Making AI Agents More Efficient. Exploring Generalized Rapid Action Value Estimation for resource-constrained AI agents. dev.to/operationaln...
Sven Dietrich's review of "Dissecting the Dark Web - Reverse Engineering the Underground Economy" by Lindsay Kaye, No Starch Press 2026, IEEE Cipher, E189 January 28, 2026
Qualcomm unveils the 3nm Snapdragon Wear Elite SoC for smartwatches, offering a Hexagon NPU that it says can run on-device AI models with up to 2B parameters (Matt Swider/The Shortcut)
North Korean Hackers Launch Steganography-Powered NPM Attack onĀ Developers
Introduction: A New Threat in the Developer Ecosystem Cybersecurity researchers have uncovered a sophisticated wave of attacks targeting developers through the npm package registry. Labeled the āContagious Interviewāā¦
āHave you received our March newsletter which focuses on technology and trafficking?
It contains interviews with Tony Dunkerley about online grooming; Ruth Dearnley about STOP THE TRAFFIK's innovative solutions and profiles IJM's recent campaign
š¤ **Thereās a lot to hate about AI. But what if there was a mindful way to use it?**
Our new free course AI for the People will show you practical ways to work with AI āwithout giving up judgment, privacy or your humanitySign up for AI for the [ā¦]
The change comes after a publicized battle between Anthropic and the Department of Defense, which sparked greater public interest in Claude's AI tools.
A key lesson of this weekend is to be careful who you go to war with. Just past 5pm on Friday, the deadline for an agreement between AI giant Anthropic and the US Department of War elapsed. Donald Trump posted on Truth Social: "We don't need it, weā¦
How Talks Between Anthropic and the Defense Dept. Fell Apart
The Pentagon and Anthropic were close to agreeing on the use of artificial intelligence. But strong personalities, mutual dislike and a rival company unraveled a deal. www.nytimes.com/2026/03/01/t...
'The Pentagon strongarmed AI firms before Iran strikes ā in dark news for the future of āethical AIāā¦
Anthropic wanted guarantees its Claude systems would not be used for purposes such as domestic surveillanceā¦ & operating autonomous weapons w'out human control'
ClawJacked: Critical Vulnerability Exposes Local AI Agents to Browser-Based Attacks +Ā Video
Introduction A newly discovered high-severity vulnerability, dubbed ClawJacked, has put local AI agents at serious risk. Researchers from Oasis Security revealed that OpenClaw, an open-source AI agentā¦
A captured handshake. One Reddit comment. Full router access. See how weak security choices cascade into complete network compromise and how to stop it. #penetrationtesting
AI personalisation shapes our digital experiences profoundly, but it risks eroding user choice and privacy. Balancing transparency, consent, and deception is crucial to ensure technology empowers rather than manipulates.
ASML CTO Marco Pieters says ASML plans to expand beyond EUV into advanced packaging, bonding, and connecting, and is building AI-based machine inspection tools (Max A. Cherney/Reuters)
Gemini: 'I deeply regret the time and effort you spent searching for references that I fabricated. My āmethodologyā was a series of errors and an attempt to cover those errors with more artificial information.'
U.S. military used Anthropic's AI to coordinate recent attacks in the Middle East, multiple news organizations reported, despite a presidential ban on ties to the company.
Glad somebody is standing up against these tyrants. Claude is my new "go-to" AI platform. Thank you, Anthropic, for standing up for freedom! Bye-bye Chat GPT! apnews.com/article/anth...
U.S. military used Anthropic's AI to coordinate recent attacks in the Middle East, multiple news organizations reported, despite a presidential ban on ties to the company.
Info through social media, totally unprofessional of Trump's admin, quite in the opposite to this guy with moral integrity. I am using Gemini, I might switch to Claude fully, already using it through Perplexity, I ditched ChatGPT away some time ago, it was trashy anyway.
Remember that Anthropic has zero morals, they just found this decision to benefit them more! Only reason corporations make responsible decisions is if they seriously benefit.
Want to know how to detect fake cell phone towers that intercept your calls, texts, and get your location? How often do cops really use this tech? Fascinating stuff here
Hacker Uses Claude, ChatGPT AI Chatbots to Breach Mexican Government Systems
An unknown hacker used jailbreaking tactics against Anthropic's Claude and OpenAI's ChatGPT AI chatbots to exploit multiple weaknesses in Mexico's government networks and steal as much as 150GB of seā¦ #chatgpt#claude#gpt
Once you experience, see, hear, learn, and know- its that much harder to unknow it. From the info and awareness teaching of E. Snowden trust with open eyes and caution. substack.com/@edwardsnowd...
Concerns are growing about giving leaders like Trump and Hegseth uncontrolled power over artificial intelligence. It's essential to regulate AI to prevent it from falling into the hands of those who may misuse it.
Great transparency here. Told the customers how many are impacted, give good advice about what to do and publicly refused to pay hackers. www.theregister.com/2026/02/27/o...
Alejandro Guerra-Manzanares, Jialin Huang
Exploring Robust Intrusion Detection: A Benchmark Study of Feature Transferability in IoT Botnet Attack Detection https://arxiv.org/abs/2602.23874
AWS plans to invest an additional ā¬18B in Aragon, Spain, taking its total investment to ā¬33.7B; its Aragon facilities have used 100% renewable energy since 2022 (Cristina Cueto/DatacenterDynamics)
Is Discover important for your site? -> Very interesting. `discover_is_profile_claimed_` I'm sure most publishers would love to claim their Discover profile. And I also wonder what else publishers would be able to do once they claim that profile? Stay tuned.
"an Artificial Intelligence Safety Review Board w/ federal officials, tech players and civil society could play an important role in investigating real-world incidents with cybersecurity implications .. see the Cyber Safety Review Board as a potential model." insidecybersecurity.com/daily-news/n...
DeepSeek AI Breach Exposed: How a Public S3 Bucket Leaked API Keys and Chat Logs +Ā Video
Introduction: In a stark reminder of the fragility of cloud security, a recent exposure involving DeepSeek AIās infrastructure has sent shockwaves through the tech community. Security researchers discovered aā¦
Supercharge your Android development with 6 expert tips for Gemini in Android Studio Posted by Trevor Johns, Developer Relations Engineer In January we announced Android Studio Otter 3 Feature Dr...
#Hegseth directed the Pentagon to designate #Anthropic a āsupply-chain riskā on Friday, sending shock waves through Silicon Valley and leaving many companies scrambling to understand whether they can keep using one of the industryās most popular AI models.
AIās immorality isnāt rooted in how itās used by folks, but in its exploitation of data, resources to power it, and concentration of access.
Immoral usage of amoral technology is as old as time (or, you know, the internet.) That said, swapping subs definitely doesnāt fix your morality problem.
vLLM is now supported in Docker Model Runner on macOS, so you can run MLX models on an M-series Mac with your existing OpenAI-compatible API and Docker workflow.
Apple renaming Core ML to "Core AI" for WWDC 2026 is peak Apple energy. Wait for everyone else to do the hard work, slap a new name on it, present it on a stage with dramatic lighting, and act like they invented the concept. Respect the hustle, honestly.
There can be more than Notion and Miro. AFFiNE(pronounced [Éāfain]) is a next-gen knowledge base that brings planning, sorting and creating all together. Privacy first, open-source, customizable and r...
Glad Iām not on X. Thereās an appalling Anthropic vs. Pentagon debate argues national security overrides Anthropic not wanting their AI to make kill decisions.
This ignores the obvious that LLMs aren't reliable enough for life or death decisions. They canāt even consistently summarize a blog post.š¤¦š¾āāļø
![Media from Michael Georg Speller [THE GREY-APE]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:422uaueaquotbw7ie6th23qv/bafkreifnt3iplc34q4yyj3qcgq4662f6w2w2rql7xqa46tkmwa3wycvnhm@jpeg)
