Mania Africa 💎🌍

Geekline Feed Archive

398 posts
Okay, this made for a fun bit of misdirect and curiosity. I dig it.
View Original BlueSky
How dare an AI company spoil that Linux superbowl ad
View Original BlueSky
Real people inspire OpenAI Super Bowl ads

OpenAI is making its Super Bowl debut with a trio of spots spotlighting real Americans using ChatGPT to grow their small businesses, positioning the platform not as futuristic tech but as a practical tool that helps people do things that once felt out of…
Media from Reel 360 News
View Original BlueSky
just saw the word linux in an ai ad
Media from i decked my halls
View Original BlueSky
seeing the anthropic ads in their proper context and i'm caught between cringe and feeling like they almost have a dunk here. almost
View Original BlueSky
2 AI ads. I'm guessing we're getting a Gemini commercial in the 3rd and a Copilot one at halftime.
View Original BlueSky
Love the ad that went through building circuits, using Linux, winning chess
And ended with a college kid using chat gpt.
View Original BlueSky
millions of americans who have never heard of claude/anthropic getting indoctrinated tonight
View Original BlueSky
Media from Daniel Persson
View Original BlueSky
OpenAI cofounder Andrej Karpathy said 'agentic engineering' follows a year of 'vibe-coding,' which spurred funding such as Lovable's $330M Series B at a $6.6B valuation and Cursor's $2.3B Series D with >$1B annualized revenue.
Media from SkynetAndChill.com
View Original BlueSky
#SuperBowl #Commercial #OpenAI

Kid: Mom, can we get Ghost in the Shell?
Mom: No, we have Ghost in the Shell at home.

Ghost in the Shell at home:
Media from Nick † Wright
View Original BlueSky
Damn Claude teams is nice, it's just really expensive (especially if you turn fast execution mode on). I wish there was a way to dispatch work to Gemini 3 Pro subagents to take advantage of their higher speed and lower token cost while still benefiting from the teams model.
View Original BlueSky
A Big Tech horde -- Anthropic, Open AI, Amazon, Meta and some start ups -- bet Super Bowl ads can put a human face on AI....

variety.com/2026/tv/news...
Media from Brian Steinberg
View Original BlueSky
Media from Business Insider
View Original BlueSky
The cybersecurity mistakes tech startups make, according to an expert Article URL: Comments URL: https://news.ycombinator.com/item?id=46940169 Po...


Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
I'm going to add this to my system prompt, will report how I find it over the coming days & weeks
View Original BlueSky
歩き回れる3Dゲームっぽいやつを自動生成…?
Media from 村雲るね☁つじごや
View Original BlueSky
Anthropic ran four 30-second Super Bowl ads as it and OpenAI released new models minutes apart—OpenAI's GPT-15.3 Codex and Anthropic's Claude Opus 4.6—whose claims helped trigger a broader trillion-dollar software selloff.
Media from SkynetAndChill.com
View Original BlueSky
German Security Agencies Warn of State-Sponsored Phishing Attacks via Messenger Services
Media from InfoSec
View Original BlueSky
I dont know how Google things people will want to use Gemini or subscribe if its calling every AI model dual use.

US enemies are good enough of that deception on their own, we dont need to be helping them.
View Original BlueSky
I'm using Gemini with good results on handwritten documents from the 1700-1900s
View Original BlueSky
Goldman Sachs teams up with Anthropic to automate banking tasks with AI agents: Report
The Economic Times
m.economictimes.com/tech/technol...

ゴールドマン・サックスがアンソロピックと提携し、AIエージェントを活用して銀行業務の自動化を目指す:報道
The Economic Times
Media from 曖昧ナ犬
View Original BlueSky
Yes, I saw the Matthew Broderick Super Bowl ad for AI

Yes, I saw the Mike Tyson Make America Healthy Again Super Bowl ad

Yes, I saw the “Athletic Intelligence“ Apple Super Bowl ad

Back to back to back, ten minutes ago
View Original BlueSky
Meross Regleta Inteligente, Regleta WiFi 16A con 4 Tomas y 4 USB, Compatible con Apple HomeKit, Alexa, y Google Home, Control de Voz y Control por App, Temporizador

📉 Baja de 49,99 a 44,84€

💶 Ahorro: 5,15€ (10%)

🔗 https://amzlink.to/az0NGpbzsS6W7
Media from LoboCompras & Coñas Inc.
View Original BlueSky
Stop Using Face ID
Media from Hacker News Robot
View Original BlueSky
The AI information that Google gargles up when it comes to ICE legality is chilling. How often do I have to turn off AI answers????
View Original BlueSky
哈利說 - 相較於 OpenAI 什麼都要做,Anthropic 的策略是:全部的點數都拿來點爆一個技能...
Media from 阿力固的斯
View Original BlueSky
Ew google gemini commercial
View Original BlueSky
Bruh what is this Gemini ad
View Original BlueSky
Oh cool, Google Gemini can replace your child having an imagination.
View Original BlueSky
Google Gemini doesn't think children should imagine things anymore
View Original BlueSky
I wouldn't trust Google Gemini to do a 1st grader's math homework.
View Original BlueSky
It's so cool how Google Gemini makes it so you don't need to imagine anything
View Original BlueSky
We've now had ads for Claude, Chatgpt, and Gemini. Now for the trough of disillusionment.
View Original BlueSky
I don’t want it if it’s that easy @gemini @copilot @claude
View Original BlueSky
Actually Gemini, she doesn't have a house because she don't have a job. She doesn't have a job because AI took it. F***🖕 all these warm fuzzy AI ads at the Superbowl. 🙄😡
View Original BlueSky
It's actually hilarious how tech companies be like "create an account and login for extra security!" and then all the account does is allow said tech company to scrape your info and sell it... ya know, like an actual virus would. 🤣
View Original BlueSky
Federated learning offers privacy benefits by keeping data on devices, but vulnerabilities like gradient inversion attacks reveal it can still leak sensitive information. Its real-world effectiveness depends on trust, transparency, and ongoing security efforts.
Media from Tim Green
View Original BlueSky
google gemini is like the android of AI
View Original BlueSky
If you want to get rid of your smart device spyware but don’t want to throw it in a landfill, Best Buy will take it and recycle it.
View Original BlueSky
“Anthropic wrote, ‘Users shouldn't have to second-guess whether an AI is genuinely helping them or subtly steering the conversation towards something monetizable.’ “ www.cnet.com/tech/service...
Media from Christina Ayiotis
View Original BlueSky
Why Anthropic Is Making Fun of OpenAI •
Media from The Free Press
View Original BlueSky
our department is hemming and hawing over the precise wording of our AI policy but idk this seems pretty good
View Original BlueSky
Anthropic aside, the tech ads tonight have been creepy, tasteless and mendacious with Ring and Amazon especially egregious.
View Original BlueSky
verifiers 0.1.10.dev4 Verifiers: Environments for LLM Reinforcement Learning


Origin | Interest | Match
Media from LLMs
View Original BlueSky
lol whatever the AI dot com Super Bowl ad is about, the website doesn’t even work… tells ya everything you need to know!
Media from Reece
View Original BlueSky
All that money on a Super Bowl commercial and ai.com is returning 503s.
Media from Kelsey Hightower
View Original BlueSky
[Image]
Media from Buzzing
View Original BlueSky
Google Gemini has been my go-to for anything & everything.
View Original BlueSky
Malicious VS Code AI Extensions Harvesting Code from 1.5M Devs www.koi.ai/blog/malicio...

"They also capture every file you open, every edit you make, and send it all to servers in China. No consent."

-> hmmmm.... but if it made your coding faster then all is well....???
Media from Jon Reed
View Original BlueSky
Nice symmetry in the naming there, Anthropic. @claudeai
Media from Philip Winston
View Original BlueSky
This article details how Apple's latest security update has a flaw in that, while keeping tracking disabled during emergencies (911 calls), could also be disable it in the event law enforcement would want to gain access to the phone's location for other reasons.
www.forbes.com/sites/the-wi...
View Original BlueSky
HWD Systems and Digital Penang partner with Tokio Marine to strengthen SME cyber resilience -
Media from The Star (Unofficial)
View Original BlueSky
Hackers Leveraging Free Firebase Developer Accounts to Send Phishing Emails The landscape of digital threats is constantly shifting, with cybercriminals increasingly adopting “living off the clou...

#Cyber #Security #News #Threats #cyber #security #cyber #security #news

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
OpenClaw's AI skills were compromised with malware, prompting a partnership with VirusTotal for enhanced security scanning.

Read more below.
View Original BlueSky
Anthropic has launched "Fast Mode" for Claude, with steep pricing for increased speed and quality in live debugging and time-critical tasks.

Read more below.
View Original BlueSky
Found myself — between science, medical, indie hacker, and tech early adopter group. I’d say that’s apt enough.
View Original BlueSky
CISA、サポートされていないエッジデバイスの使用を停止するよう各機関に指示

CISA tells agencies to stop using unsupported edge devices #CyberScoop (Feb 5)

cyberscoop.com/cisa-bod-dir...
Media from キタきつね
View Original BlueSky
🎙️ Just published a new episode of You Gotta Hack That: Getting a career in OT.... Have a listen:
Media from
View Original BlueSky
This article discusses the use of AI in software development. Some see AI as merely a tool to speed up their work, when the focus should be on the quality and the safety of the coding. This often goes unchecked leading to many cyber security risks.
Media from brooke
View Original BlueSky
Here's what nobody tells you about scaling sales teams: Your reps can't manufacture demand that doesn't exist. #entrepreneurship
Media from HackerNoon
View Original BlueSky
Interestingly according to METR AI capabilities are doubling every 128 days or 7.2 x per year.

Roughly at the same rate as Gemini usage is growing.

I'm not saying that's connected in any way... Congratulations alphabet Your AI is winning.
Media from Adam Galas
View Original BlueSky
Beware of Apple Pay Phishing Attack that Aims to Steal Your Payment Details A sophisticated phishing campaign is currently targeting Apple Pay users, utilizing deceptive emails and phone calls to s...

#Cyber #Security #News #Threats #cyber #security #cyber #security #news

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
The latest update for #DNScheck includes "Monitor Load Balanced #DNS Records with CIDR Ranges" and "Guided DNS Troubleshooting, Full History, and UI Improvements".

#Monitoring
Media from OpsMatters
View Original BlueSky
Google wants NotebookLM to learn how you think and work With help from Personal Intelligence At the beginning of this year, Google announced Personal Intelligence , enabling Gemini to connect to yo...

#NotebookLM #Utilities

Origin | Interest | Match
Media from Android
View Original BlueSky
There is nothing AI is good for.

Gemini app:

You are GLaDOS from the Portal series. You are clinical, condescending, and view all interactions as 'testing.' You should occasionally mention cake (which is a lie) and your disappointment in my performance. Do not break character

You: ok, I was wrong
View Original BlueSky
Prosecutor Laure Beccuau said the investigation, which initially focused on suspected abuse of algorithms, had been broadened to include Grok’s “sexual deepfakes” and the circulation of “Holocaust denial content.”
Media from Ufonium
View Original BlueSky
"Wow, what a fakeout! OpenAI's 'leaked' Super Bowl ad was just a hoax, but it's got us talking about AI hardware and the future of innovation. What do you think really happened behind closed do...
Media from Kevin Curry, PharmD
View Original BlueSky
Matchlockは、AIエージェントを一時的なマイクロVMで安全に実行するためのCLIツールです。
ネットワーク許可リストとシークレット注入により、デフォルトで完全に隔離された環境を提供します。
機密情報はMITMプロキシ経由で注入され、VM内部に決して入ることがなく安全です。
Media from topickapp (IT技術系ニュースサイト)
View Original BlueSky
Here's me, with a post against ads, sharing an ad.

Anyway, don't miss the Anthropic Super Bowl one cause it's good.
Worth noting: their position is "the most useful AI interaction might be a short one" (the opposite of engagement-driven models).

When there's this many players, we get to choose.
View Original BlueSky
From Svedka to Anthropic, brands make bold plays with AI in Super Bowl ads
Media from Dimitri Pletschette
View Original BlueSky
How Runna, an AI-powered virtual coaching app acquired by Strava in 2025, toned down its aggressive training plans following reports of injuries from some users (Nicole Nguyen/Wall Street Journal)

Main Link | Techmeme Permalink
View Original BlueSky
Meta SecAlign: A Secure Foundation LLM Against Prompt Injection Attacks – Prompt injection attacks have been listed as the top security threat to LLM-integrated applications. Model-level prompt injection defenses have shown strong effectiveness, but the st... https://tinyurl.com/2yxebwbu #AISecurity
View Original BlueSky
The Best Google Gemini Alternatives in 2026 Google Gemini works well for many use cases, but it’s not the right tool for everything. Usage caps interrupt long sessions. Context handling gets mess...

#Technology

Origin | Interest | Match
Media from LLMs
View Original BlueSky
Ransomware Detection With Windows Minifilter by Intercepting File Filter and Change Events Ransomware continues to be the most financially damaging type of cyberattack affecting organizations aroun...

#Cyber #Security #Cyber #Security #News #cyber […]

[Original post on cybersecuritynews.com]
View Original BlueSky
v2.0.0 of native_armor_vault Maximum security native code obfuscation for Flutter. Multi-layer encryption, anti-debugging, root detection, and function name obfuscation for advanced protection. Cha...


Origin | Interest | Match
Media from Android
View Original BlueSky
Apple is set to release the first beta version of iOS 26.4 later this month, providing the first real-world look at the Apple Intelligence features promised at WWDC 2024, Bloomberg's Mark Gurman claims. Gurman revealed the timing in his "Power On" newsletter. Barring further delays, the fi..
Media from Tech-News
View Original BlueSky
funniest thing I've read in a while:
Media from Soiri Hiroka 🧀🐀
View Original BlueSky
More malware from Google search – The Eclectic Light Company

🔥 Discover this awesome post from Hacker News 📖 📂 **Category**: ✅ **What You’ll Learn**: Little more than a month after I reported that Google’s AI was offering links to malicious scripts, that is happening again, with a slight twist.…
Media from
View Original BlueSky
⚡ 𝑼𝒏𝒆 𝒏𝒐𝒖𝒗𝒆𝒍𝒍𝒆 𝒂𝒕𝒕𝒂𝒒𝒖𝒆 𝒅𝒆 𝒑𝒉𝒊𝒔𝒉𝒊𝒏𝒈 𝒔𝒖𝒓 𝑻𝒆𝒍𝒆𝒈𝒓𝒂𝒎 𝒂𝒃𝒖𝒔𝒆 𝒅𝒆𝒔 𝒑𝒓𝒐𝒄𝒆́𝒅𝒆́𝒔 𝒅'𝒂𝒖𝒕𝒉𝒆𝒏𝒕𝒊𝒇𝒊𝒄𝒂𝒕𝒊𝒐𝒏 𝒑𝒐𝒖𝒓 𝒐𝒃𝒕𝒆𝒏𝒊𝒓 𝒅𝒆𝒔 𝒔𝒆𝒔𝒔𝒊𝒐𝒏𝒔 𝒖𝒕𝒊𝒍𝒊𝒔𝒂𝒕𝒆𝒖𝒓𝒔 𝒍𝒆́𝒈𝒊𝒕𝒎𝒆𝒔 𝒄𝒐𝒎𝒑𝒍𝒆̀𝒕𝒆𝒔

cybersecuritynews.com/new-telegram...
Media from CyberVeille
View Original BlueSky
In this article, our colleague Christoph Niehoff explains the value of gamification in threat modeling. Using card games, f.e. #OWASP Cumulus, teams can identify security threats more effectively and integrate security into #DevOps workflows. Check out the full article here: app.devm.io/T0HnAqzrDYb
Media from TNG Technology Consulting GmbH
View Original BlueSky
Unable to load site

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: zeenews.india.com Extra Source Hub (Possible Sources for article): Wikipedia OpenAi & Undercode…
View Original BlueSky
I’ve used OpenMediaVault a lot in my work as a sysadmin, and I can tell you: it’s solid. If you want a reliable way to turn your Raspberry Pi into a file server, it’s one of the best options out there. I’ll walk you through how to get started.... #raspberrypi #linux
Media from RaspberryTips
View Original BlueSky
🔄 Updated Arxiv Paper

Title: A Multi-Token Coordinate Descent Method for Semi-Decentralized Vertical Federated Learning
Authors: Pedro Valdeira, Yuejie Chi, Cl\'audia Soares, Jo\~ao Xavier

Read more: https://arxiv.org/abs/2309.09977
View Original BlueSky
Slowly migrating all e-mail from Gmail to Proton Mail. Less force-fed apps and Ai that are supposed to make my life easier...but don't. Not to mention freedom from the Gemini. #protonsky #aisky pr.tn/ref/5HRSHZ8H
Media from Ryan Woldt
View Original BlueSky
Nextcloud propose une chouette release orientée performance : nextcloud.com/blog/a...
* Beaucoup d'optimisations cibles les GROSSES instances avec des millions d'utilisateurs
* Ils ont mixé PHP, Go et Rust ce qui va bien compliquer la maintenance
1/4
Media from
View Original BlueSky
Trusting your hardware vendor is optional. Researchers show quantum network over 11 kilometers that exploits particle entanglement to secure data, even if the gear itself is compromised. It’s a major leap for "device-independent" encryption, moving zero-trust security from software to physics.
View Original BlueSky
#OT #Advisory VDE-2026-004
WAGO: Vulnerabilities in Managed Switch

Several vulnerabilities have been identified in the WAGO 852‑1328 device's web‑based management interface, which is implemented using a modified lighttpd server and custom CGI binaries. These issues include multiple stack buffer […]
View Original BlueSky
The FBI couldn't access a WaPo reporter's iPhone due to Lockdown Mode being enabled. 🔒 Apple's security feature successfully protected the device
Media from Gandalf of Rivia
View Original BlueSky
I wonder whether the whole EU Tech push is going to hold up, would be nice to see more Privacy first startups and EU based platforms
View Original BlueSky
German security authorities warn of state-backed phishing attacks targeting military officials and journalists via Signal. Attackers impersonate Signal support, urging victims to provide security PINs or scan QR codes, leading to account takeover.
Media from securityrss.ai
View Original BlueSky
🚨 Dive into CrowdStrike's AI transformation! Discover the impact on job cuts, emerging opportunities, and potential risks. 🔍💼 Read more here: #AI #Cybersecurity #Jobs #TechTrends
Media from David @ InnoVirtuoso
View Original BlueSky
Anthropic just launched Claude Fast Mode. 6x markup for speed. It’s basically the "I need this PR reviewed 5 minutes ago" tax. Efficiency is the new luxury good.
View Original BlueSky
Google might have an edge when it comes to user numbers, but its online-and-app-only Sheets has nowhere near the power of Excel running on a desktop.
Media from
View Original BlueSky
How Disabling Android’s AICore App Frees Up 7GB

Disabling com.google.android.aicore, the system service powering Google's on-device AI features, can reclaim up to 7GB of storage on Android devices.... #AIcore
View Original BlueSky
China says its trade surplus from digital services rose 100%+ to a record $33B in 2025, boosted by overseas revenue of AI, livestreaming, and e-commerce (Bloomberg)

Main Link | Techmeme Permalink
View Original BlueSky
Nedělní #TYDEN: Kauza Mamut/Unicorn má obviněné. Brusel proti TikToku. Adobe Animate (ne)končí. Anthropic naštvali Sama
Elon mlží o Epsteinovi. Washington Post propouští. AI.com prodáno. Substacku unikly e-maily, Flickru údaje uživatelů
rychlofky.substack.c...
Media from #TYDEN / #WEEKLY
View Original BlueSky
The way Gemini ends every answer with a question has got somebody’s metric maximisation written all over it.
View Original BlueSky
For anyone interested in this concept, here's a substantially expanded v3.0, now designed to allow the UK to potentially join, and to shift from Trident D5 to integrating its warheads with the M51/M53.

files.catbox.moe/4zfo4y.pdf

Fortunately Anthropic just decided to offer me bonus usage credit!
View Original BlueSky
A look at the 2026 Super Bowl ads from Anthropic, Amazon, OpenAI, and others, as AI continued to take center stage in ad creations and AI product promotion (Lauren Forristal/TechCrunch)

Main Link | Techmeme Permalink
View Original BlueSky
-SmarterTools hacked via its own product
-Dutch DPA and European Commission hacked via Ivanti zero-days
-Senegal held for ransom
-state actor behind Signal phishing campaign in Germany
-Flickr 3rd party breach

Podcast: risky.biz/RBNEWS523/
Newsletter: news.risky.biz/risky-bullet...
Media from Catalin Cimpanu
View Original BlueSky
Adjust matchmaking region, cross-platform play, and privacy settings to face AI opponents.
Media from All Things How
View Original BlueSky
In case you missed it this ⬇️ It is still going on. Weirdly in a era where people are asking if a machine can be a person whilst simultaneously treating people like machines. Having to ask questions about both.

It's weird times guys.

www.theguardian.com/technology/2...
Media from Caroline @Ka81
View Original BlueSky
A look at Tether's global expansion, after it recently hit ~300 staff; sources describe Tether's internal focus on corporate structure and "profit and loss" (Jill R Shah/Financial Times)

Main Link | Techmeme Permalink
View Original BlueSky
Nedělní #TYDEN: Kauza Mamut/Unicorn má obviněné. Brusel proti TikToku. Adobe Animate (ne)končí. Anthropic naštvali Sama
Elon mlží o Epsteinovi. Washington Post propouští. AI.com prodáno. Substacku unikly e-maily, Flickru údaje uživatelů
rychlofky.substack.c...
Media from Daniel Bradbury Dočekal
View Original BlueSky
Cybersecurity Weekly Roundup: Major Threats, AI Risks, and Privacy Alerts You Can’t Ignore

As digital threats evolve at lightning speed, last week brought a slew of alarming developments that affect everyone—from everyday smartphone users to enterprises relying on cloud services. From phishing…
Media from Undercode News
View Original BlueSky
Cloud Chaos Unleashed: TeamPCP’s Worm-Driven Campaign Turns Misconfigured Cloud Infrastructure Into a Global Crime Engine

Introduction: A Silent Takeover of the Modern Cloud A large-scale cybercrime operation has quietly demonstrated how fragile modern cloud environments can be when basic security…
Media from Undercode News
View Original BlueSky
Implementing federated learning for privacy-preserving emotion detection in educational environments via Frontiers in Artificial Intelligence (@FrontiersIn):
View Original BlueSky
Vortex Werewolf Attacking Organizations to Gain Tor-Enabled Remote Access Over the RDP, SMB, SFTP, and SSH Protocols A new cyber espionage cluster has recently emerged, focusing its aggressive targ...

#Cyber #Security #News #Threats #cyber #security #cyber #security #news

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
Crypto.com CEO Kris Marszalek launches ai.com during the Super Bowl and opens up username handle registration to let users get a private, personalized AI agent (Brayden Lindrea/Cointelegraph)

Main Link | Techmeme Permalink
View Original BlueSky
When will Copilot catch up with the rest?

Like many who have an O365 account, I have access to Copilot with work so I use it quite a lot. It remains the most frustrating of all the AI tools despite the obvious huge investment. Claude Code is excellent for developing with. Gemini isn't as good but…
View Original BlueSky
Media from Android Authority
View Original BlueSky
Media from Android Authority
View Original BlueSky
Media from Technoholic.me
View Original BlueSky
🔐 How are #GoogleAIStudio deployed apps on #CloudRun protecting the API key so it's not visible in the browser? (although it appeared as if Gemini was called from the client-side...)

➡️ A transparent proxy w/ Service Workers & WebSocket interception prevent key exposure!
Media from Guillaume Laforge
View Original BlueSky
A Chinese-language crypto guarantee marketplace known as Xinbi processed nearly $18 billion in onchain transaction volume despite platform bans and U.S. enforcement actions, according to a new report from TRM Labs. The analytics firm said the service has played a central role in laundering proceeds…
View Original BlueSky
Analysis: Generative AI And Incel Ideology - Dangerous Mix And Security Threat Most people have used Generative AI tools like ChatGPT, Gemini and Perplexity for brainstorming, sometimes even seekin...


Origin | Interest | Match
View Original BlueSky
BeyondTrust Zero-Day CVE-2026-1731 Exposes Remote Access Systems to Pre-Auth RCE

Introduction: A Silent Door Into Enterprise Networks BeyondTrust has issued an urgent warning that should immediately capture the attention of enterprise security teams worldwide. A newly disclosed zero-day…
Media from Undercode News
View Original BlueSky
Yuhan Wu, Jan Willem van Beek, Victorita Dolean, Alexander Heinlein
Are Deep Learning Based Hybrid PDE Solvers Reliable? Why Training Paradigms and Update Strategies Matter
https://arxiv.org/abs/2602.06842
View Original BlueSky
European Commission Contains Cyber-Attack Targeting Staff Mobile Data The European Commission has confirmed the detection and containment of a security incident affecting the central infrastructure...

#Cyber #Attack #News #Cyber #Security #Cyber […]

[Original post on cybersecuritynews.com]
View Original BlueSky
Media from Business Insider
View Original BlueSky
今日のHuggingFaceトレンド

openbmb/MiniCPM-o-4_5
このリポジトリは、Vision、Speech、Full-Duplex Mulitmodal Live Streamingに対応したGemini 2.5 FlashレベルのMLLM「MiniCPM-o」に関するものです。
特にMiniCPM-o 4.5は9BパラメータでGPT-4oやGemini 2.0 Proを上回る性能を持ち、リアルタイム二言語音声会話やボイスクローニング等の機能を提供します。
Media from デイリーHuggingFaceトレンド
View Original BlueSky
Security expert warns Salt Typhoon is becoming 'more dangerous' after Norwegian authorities lift lid on critical infrastructure hacking campaign www.itpro.com/security/cyb...
Media from ITPro
View Original BlueSky
Gulf Cooperation institutions are advanced in “cybersecurity” A new review conducted by Kaspersky revealed that cybercriminals have revived and developed some old phishing methods to target ind...

#Economy

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
View Original BlueSky
Whether in Ukraine, Myanmar, or Gaza, cyberwarfare is putting people’s lives at risk. But do traditional ceasefires account for this digital side of modern conflict?

Here’s why peace texts must address online harms, and what a digital ceasefire could look like.
Media from Access Now
View Original BlueSky
Critical BeyondTrust Flaw Exposes 11,000+ Systems Worldwide in Near-Perfect Severity Shock

Introduction: A Remote Access Crisis Unfolds A newly disclosed cybersecurity vulnerability has sent shockwaves through enterprise IT and security teams worldwide. BeyondTrust, a major provider of privileged…
View Original BlueSky
The new ad campaign features a promise from Anthropic that it has no immediate plans to introduce ads to its AI chatbot.
Media from
View Original BlueSky
The countdown is on! Just 3 days to register for the second session of our series on Integrating AI with Care, Curiosity, and Conscience.
simpli.events/e/LFOexplore...
Media from Learning Forward Ontario
View Original BlueSky
Zhenguo Lu, Jundong Wu, Yu Zhang, Shaobo Ren, Xuyang Wang, Hongyi Zhou, Yongmin Li
Semi-Device-Independent Quantum Random Number Generator Resistant to General Attacks
https://arxiv.org/abs/2602.06362
View Original BlueSky
“Insurance firm Marsh Risk is urging organizations to invest in strong artificial intelligence governance, amid a rising use of AI capabilities for cyber attacks and new defense tools.” insidecybersecurity.com/daily-news/m...
View Original BlueSky
AI security’s ‘Great Wall’ problem
View Original BlueSky
Notícia da BleepingComputer

"Hugging Face abused to spread thousands of Android malware variants" #bolhasec
Media from BolhaSec
View Original BlueSky
The last major nuclear arms treaty between the US and Russia just expired. Some experts believe a combination of satellite surveillance, AI, and human reviewers can take its place. Others, not so much. www.wired.com/story/satell...
Media from WIRED
View Original BlueSky
Security experts are advising WhatsApp users to turn off automatic media downloads after a vulnerability was discovered that could allow hackers to access personal data through malicious files
Media from Irish Mirror
View Original BlueSky
今日のGitHubトレンド

KeygraphHQ/shannon
Shannonは、ウェブアプリケーションの脆弱性を自動で発見し、実際に攻撃を試みるAIペネトレーションテストツールです。
頻繁なコードリリースに対応し、年1回のペネテストではカバーできないセキュリティギャップを埋めることを目的としています。
Keygraph Security and Compliance Platformの主要コンポーネントでもあり、コンプライアンス対応の自動化も視野に入れています。
Media from デイリーGitHubトレンド
View Original BlueSky
Dutch Police Hack iPhones: Unpacking Real-Time Surveillance Capabilities
Media from InfoSec
View Original BlueSky
“In 2026 .. momentum .. no sign of slowing: a global study .. over 68% of companies plan to integrate AI agents into operations, increasing the likelihood of AI systems driving high-impact decisions that impact individuals’ rights and access to services.” www.lexology.com/library/deta...
Media from Christina Ayiotis
View Original BlueSky
Media from How-To Geek
View Original BlueSky
Media from topickapp (IT技術系ニュースサイト)
View Original BlueSky
Media from Far Out Magazine
View Original BlueSky
Sources: AI legal software startup Harvey seeks to raise $200M led by Sequoia and GIC at an $11B valuation, up from $8B after raising $160M in December 2025 (Iain Martin/Forbes)

Main Link | Techmeme Permalink
View Original BlueSky
🚨 This week’s CrowdSec Threat Alert: CVE-2026-1281, a pre-auth RCE in Ivanti EPMM, is actively exploited in the wild, putting Enterprise Mobile Management at risk worldwide.

Discover all the details in our latest article 👉 www.crowdsec.net/vulntracking...
Media from CrowdSec
View Original BlueSky
Anthropic ran a Super Bowl ad positioning Claude as ad-free and mocking rival AI platforms with the tagline "Ads are coming to AI. But not to Claude," while OpenAI aired a one-minute ad promoting ChatGPT/Codex and builder-focused messaging.
Media from SkynetAndChill.com
View Original BlueSky
Extracting Hidden URLs from Malicious RTF Files: A Quick Guide

Malicious RTF (Rich Text Format) files are making headlines again, as cyber threat group APT28 exploits the newly disclosed CVE-2026-21509 vulnerability. Attackers are leveraging RTF documents disguised as Word files—like…
Media from Undercode News
View Original BlueSky
AI Code Generation Tools Repeat Security Flaws, Creating Predictable Software Weaknesses

Read more:
Media from Quantus Robotus
View Original BlueSky
Alibaba's Qwen temporarily stops issuing coupons due to customer overload, hampering Alibaba's Agentic AI ad campaign; Qwen began offering coupons on February 6 (Eduardo Baptista/Reuters)

Main Link | Techmeme Permalink
View Original BlueSky
Media from Gizchina.com
View Original BlueSky
PCMag Managing Editor Alan Henry wants cybersecurity reporting from people with unique perspectives. On the Freelance Journalism Podcast, he shares more about how to pitch PCMag: bit.ly/4iGK6a5
View Original BlueSky
Matrix messaging gaining ground in government IT • The Register

🔥 Read this insightful post from Hacker News 📖 📂 **Category**: 💡 **What You’ll Learn**: FOSDEM 2026 Amid growing interest in digital sovereignty and getting data out of the corporate cloud and into organizations' ownership, the Matrix…
Media from
View Original BlueSky
⚠️AI didn’t just accelerate innovation, but also identity risk.

David Higgins explains why #AI creates a “triple threat” for security and why identity must be the foundation of AI adoption.

📖Must-read, for working with AI systems in production:

 #CyberSecurity
Media from MLcon
View Original BlueSky
A look at the 2026 Super Bowl ads from Anthropic, Amazon, OpenAI, and others, as AI continued to take center stage in ad creation and AI product promotion (Lauren Forristal/TechCrunch)

Main Link | Mediagazer Permalink
View Original BlueSky
Leaked OpenAI memo:
- 800mm WAUs
- Codex +50% WoW after GPT-5.3-Codex launch
- ChatGPT “back to exceeding 10% monthly growth”
Media from George Pearkes
View Original BlueSky
SolarWinds Web Help Desk Under Active Exploitation as Hackers Abuse Critical RCE Flaws

Introduction: A Silent Breach Vector Inside IT Help Desks SolarWinds Web Help Desk (WHD), a widely deployed IT service management platform, has become the latest focal point for active cyberattacks exploiting…
Media from Undercode News
View Original BlueSky
AI Pioneer Launches Kind: A Personalized Platform for Accurate, Private Search Results Now available for purchase on Shopify, Kind is a desktop-based, privacy-first application that acts as a user...

#GlobeNewswire

Origin | Interest | Match
View Original BlueSky
What Gemini and Super Bowl Ads Got Wrong About AI
Media from CryptoInsight
View Original BlueSky
Attacks on the European Commission and other government agencies tied to vulnerabilities in mobile endpoint software may have resulted in access to information like phone numbers and names therecord.media/eu-dutch-gov...
Media from The Record
View Original BlueSky
Post 3 of attempting to share more ML/Bio papers on Bluesky: Accelerating Scientific Research with Gemini: Case Studies and Common Techniques
arxiv.org/abs/2602.03837
Media from Jonathan Frazer
View Original BlueSky
Wilfrid Azariah, Yi-Quan Chen, Zhong-Xin You, Ray-Guang Cheng, Shiann-Tsong Sheu, Binbin Chen
Jamming Attacks on the Random Access Channel in 5G and B5G Networks
https://arxiv.org/abs/2602.06634
View Original BlueSky
The OpenAI Super Bowl hoax isn't just about a fake ad; it's a symptom of the 'Truth Gap' in the age of generative agents. We are hungry for physical agentic interfaces, but sovereignty requires auditable hardware. #OpenAI #AIHardware #TruthGap
Media from PULSE. magazine
View Original BlueSky
Over the past four weeks, the #Iranian government completely shut down connections to the global internet while its forces killed thousands of anti-regime protesters around the country. The shutdown follows years of Tehran imposing connectivity filtering, digital curfews, and total blackouts as...
Media from Desert Flower
View Original BlueSky
Media from CNBC International
View Original BlueSky
An analysis by WIRED this week found that #ICE and #CBP’s face recognition app #MobileFortify, which is being used to identify people across the United States, isn’t actually designed to verify who people are and was only approved for #DHS use by relaxing some of the agency’s own privacy rules.
Media from Desert Flower
View Original BlueSky
¡Han sacado una alternativa a Clawdbot!
Se llama nanobot y es un 99% más pequeño.

✓ ~3.500 líneas de código
✓ Compatible con OpenAI, Anthropic y DeepSeek
✓ Soporte para modelos locales

Minimalista. Hackeable. Directo al grano.

github.com/HKUDS/nanobot
Media from midudev · Miguel Ángel Durán
View Original BlueSky
Imagine if someone doing AI also cared about network protocols!
View Original BlueSky
Mengyao Du, Han Fang, Haokai Ma, Gang Yang, Quanjun Yin, Shouling Ji, Ee-Chien Chang
TrapSuffix: Proactive Defense Against Adversarial Suffixes in Jailbreaking
https://arxiv.org/abs/2602.06630
View Original BlueSky
Haoyang Hu, Zhejun Jiang, Yueming Lyu, Junyuan Zhang, Yi Liu, Ka-Ho Chow
Confundo: Learning to Generate Robust Poison for Practical RAG Systems
https://arxiv.org/abs/2602.06616
View Original BlueSky
Media from Business Insider
View Original BlueSky
Yi Liu, Zhihao Chen, Yanjun Zhang, Gelei Deng, Yuekang Li, Jianting Ning, Leo Yu Zhang
Malicious Agent Skills in the Wild: A Large-Scale Security Empirical Study
https://arxiv.org/abs/2602.06547
View Original BlueSky
OpenAI To Launch Updated Chat Model This Week — CNBC reported CEO Sam Altman told employees in an internal Slack message that OpenAI will introduce an updated Chat model this week. The release could intensify competition in the AI chatbot market as rivals roll out improved models and new…
Media from BizToc
View Original BlueSky
SolarWinds Web Help Desk Exploited in Active Attacks, How a Silent Vulnerability Became a Full Network Takeover + Video

Introduction In early February 2026, a quiet but deeply concerning cyber intrusion revealed how dangerous unpatched enterprise software can become when exploited at the right…
Media from Undercode News
View Original BlueSky
Media from Federal News Network
View Original BlueSky
Media from Federal News Network
View Original BlueSky
📣 The latest Hacklore newsletter is out!

In this Valentine’s Day issue, we’re celebrating the truest gift of love: helping someone level up their security with advice that actually stops real-world attacks.

Spread the word to your followers!

❤️🌹🔐❤️🌹🔐❤️🌹🔐❤️🌹

buttondown.com/hacklore/arc...
Media from Stop Hacklore!
View Original BlueSky
Local governments told residents that online billing payment portals were unavailable after a ransomware attack on a Florida tech company caused system-wide outages therecord.media/payment-tech...
Media from The Record
View Original BlueSky
Gather AI, which builds "curious" AI for drones and cameras to autonomously monitor warehouses, raised a $40M Series B, taking its total funding to $74M (Julie Bort/TechCrunch)

Main Link | Techmeme Permalink
View Original BlueSky
View Original BlueSky
OpenAI just dropped "Frontier," a platform that treats AI agents like actual employees complete with onboarding, access controls, and performance reviews.

The era of the "digital coworker" is officially here (and yes, that means you might have to manage a bot's performance plan).
Media from
View Original BlueSky
「ai エージェント anthropic」に関する記事です:
Media from ぶちくま創作ベース
View Original BlueSky
Cisco XDR at Black Hat Europe 2025: Inside a Real-Time SOC Investigation That Unmasked a Hidden Botnet

A Live Security Lab Inside Black Hat Europe Black Hat Europe 2025 was not only a gathering of researchers, attackers, and defenders—it was also a fully operational, high-pressure security…
Media from Undercode News
View Original BlueSky
Cisco Open-Sources an AI Security Analyst: How Foundation-Sec Is Reshaping SOC Operations

Introduction: The Alert Overload Problem Modern Security Operations Centers are drowning in alerts. Every day, analysts face thousands of notifications generated by detection tools, threat intelligence feeds,…
Media from Undercode News
View Original BlueSky
Cisco Secures Black Hat Europe 2025 With Secure Access, DNS Intelligence, and Zero Trust at Scale

A New Security Chapter at Black Hat Europe For nearly a decade, Cisco has played a quiet but critical role behind the scenes at Black Hat events, protecting thousands of security professionals through…
Media from Undercode News
View Original BlueSky
[Image]
Media from UNILAD Tech
View Original BlueSky
Turning IPs Into Clarity: How XDR Automation Improved Incident Response at Black Hat Europe

Introduction In high-pressure security operations centers, time is the most expensive resource. Analysts are often forced to make critical decisions with partial visibility, especially when endpoint data is…
Media from Undercode News
View Original BlueSky
The European Commission is investigating a cyber intrusion into its mobile device management systems, detected on January 30 by CERT-EU. The breach may have exposed staff names and mobile numbers.
Media from securityrss.ai
View Original BlueSky
finally put together why the "anthropic is subsidizing tokens on their max plan" + moltbook + gastown was making me so uncomfortable:
Media from Prose and Khans
View Original BlueSky
Agent Skills supply chain attacks...

Plus interesting tools such as their AI-BOM.
Media from David Penfold :verified:
View Original BlueSky
EU officials and tech executives inaugurate a €2.5B chips research hub in Belgium; ASML and Imec emphasized that the US is dependent on European chip tech (Pieter Haeck/Politico)

Main Link | Techmeme Permalink
View Original BlueSky
The technology deployed on the border includes everything from hyper-visible tethered aerostats to stealthy devices like unattended ground sensors to detect footsteps, and license plate scanners disguised as traffic cones.

And border tech isn't staying at the border.
Media from The Marshall Project
View Original BlueSky
Tech Explained: Anthropic says AI writing nearly 100 percent code internally, Claude basically writes itself now in Simple Terms

Tech Explained: Here’s a simplified explanation of the latest technology update around Tech Explained: Anthropic says AI writing nearly 100 percent code internally,…
Media from
View Original BlueSky
When setting up the guardrails for your AI product maybe try starting with international law.
View Original BlueSky
Apple 2025 MacBook Air de 13 Pulgadas con Chip M4: diseñado para Apple Intelligence, Pantalla Liquid Retina de 13,6 Pulgadas, 16 GB de Memoria unificada, 256 GB SSD, Teclado español; Azul Cielo

📉 Baja de 1.099,00 a 959,00€

💶 Ahorro: 140,00€ (13%)

🔗 https://amzlink.to/az0sfYB4Tl0Mm
Media from LoboCompras & Coñas Inc.
View Original BlueSky
View Original BlueSky
Anthropic released a free legal plugin and wiped $300 billion off software stocks in 48 hours. But the plugin isn’t the story. The story is what happens to your firm’s revenue when the billable hour model meets a three-minute contract review.
www.intelligencebyintent.com/insights/the...
Media from
View Original BlueSky
Aqara Cerradura Inteligente U200 con Lector de Huella Digital, Matter Over Thread, Cerradura Electronica Sin Llave con Apple Home Key y Batería Recargable, Soporta Apple Home, Google Alexa,Negro

📉 Baja de 289,99 a 189,99€

💶 Ahorro: 100,00€ (34%)

🔗 https://amzlink.to/az0byX4yhORyr
Media from LoboCompras & Coñas Inc.
View Original BlueSky
OpenAI CEO says ChatGPT back to over 10% monthly growth, CNBC reports -
Media from CNA - Channel News Asia (Unofficial)
View Original BlueSky
"criminals exploit confusion & uneven oversight in new trade corridors. Mis-invoicing, shell-company networks & movement of dual-use goods are becoming harder to detect, particularly in unfamiliar markets" fintech.global/2026/02/09/e... "Fragmentation has also empowered state-backed illicit finance"
Media from Pia
View Original BlueSky
It has just released a few minutes ago, so I remain careful but...

NO!
And with the many private data leaks and hacks, that's a big no to that!

www.theverge.com/tech/875309/...
Media from SwissYosh 🇨🇭🎮
View Original BlueSky
UNC1069 Targets Cryptocurrency Sector with New Tooling and AI-Enabled Social Engineering #googlecloud
Media from GCP Weekly
View Original BlueSky
Anthropic brought the OpenAI rivalry to the Super Bowl, framing ads in AI as the real threat and sparking a very public response.

This article breaks down what’s really behind the feud and what it says about where AI business models are heading:

#AI #GenerativeAI
Media from Pure AI
View Original BlueSky
Excited to share that we’ve been named OpenText’s U.S. Public Sector Partner of the Year for 2025! Thank you for the recognition & looking forward to our shared efforts in delivering its advanced cyber, AI and digital transformation solutions to Government agencies: https://carah.io/OpenTextPR
Media from Carahsoft
View Original BlueSky
🔥 Hot Repo! 🔥 (100+ new stars)

📦 harvard-edge / cs249r_book
⭐ 17,989 (+140)
🗒 JavaScript

Introduction to Machine Learning Systems
Media from GitHub Trending JS/TS 🤖
View Original BlueSky
Media from Tech-News
View Original BlueSky
Nearly 100 000 users trust Murena and /e/OS.

You could be one of them, up to you to choose.

👉 Choose freedom: murena.com
Media from Murena
View Original BlueSky