Mania Africa 💎🌍

Geekline Feed Archive

360 posts
Google introduces a 24-hour wait and biometric/PIN confirmation before sideloading apps from unverified developers to combat malware and scams. New verification rules support hobbyists. #AndroidSecurity #Perseus #Italy
Media from Cybersecurity News Everyday
View Original BlueSky
Apple Issues Warning: 270 Million iPhones Vulnerable to DarkSword Spyware #gadget
Media from RS Web Solutions
View Original BlueSky
OpenAI Plans Launch of Desktop 'Superapp' Article URL: https://www.neowin.net/news/openai-to-merge-atlas-browser-chatgpt-and-codex-into-a-single-desktop-super-app/ Comments URL: https://new...


Origin | Interest | Match
View Original BlueSky
Okay, This is better then nothing. It's annoying, But at least you can install third party apps not signed off by Google.
Media from Hēgweard Tsulalgi
View Original BlueSky
A Potential Breach of an Anonymous Tip App Could Have Exposed Sensitive Student Data:
Media from Education Week
View Original BlueSky
Microsoft Intune: Lock it down, warn feds after Stryker • The Register
Media from Ron Dyck
View Original BlueSky
A critical Langflow flaw (CVE-2026-33017) allows unauthenticated RCE via exec() on public endpoints. Exploits appeared within 20 hours of disclosure, affecting versions up to 1.8.1. Fixed in 1.9.0 dev release. #LangflowBug #RemoteExec #TechAlert
Media from Cybersecurity News Everyday
View Original BlueSky
Filing: Anthropic says it cannot manipulate Claude once the military has deployed it, denying DOD accusations that Anthropic could tamper with models during war (Paresh Dave/Wired)

Main Link | Techmeme Permalink
View Original BlueSky
🚨 Another supply chain attack:

Attackers republished 29 legitimate npm packages with a backdoor using compromised publisher access, enabling further propagation via stolen tokens and payload delivery through an ICP canister.

Details: socket.dev/blog/caniste...
#NodeJS
Media from Socket
View Original BlueSky
AI systems are getting better at answering questions.

But they’re still bad at crediting where information comes from.

That’s not a small issue — it directly affects the sustainability of journalism.

www.niemanlab.org/2026/03/chat...

#AI #Journalism
Media from Tong Li
View Original BlueSky
I'm setting up a new app with an AI component. I've instituted a "No Fascist, No Scab" policy.
Media from Lou Plummer 
View Original BlueSky
Trivy’s GitHub Actions were breached via 75 force-pushed tags, injecting a Python infostealer that exfiltrates CI/CD secrets and developer tokens. Attack linked to hackerbot-claw and TeamPCP. #DevOpsSecurity #GitHubAttack #TeamPCP
Media from Cybersecurity News Everyday
View Original BlueSky
Anthropic on Thursday released Claude Code Channels, a research preview feature that allows developers to control their AI coding sessions directly from messaging apps like Telegram and Discord, marking a move that puts the company in direct competition with the OpenClaw.
Media from S2udios.com
View Original BlueSky
NIST NCCoE publishes six final 5G cybersecurity guides to address critical infrastructure risks beyond network interfaces - Industrial Cyber industrialcyber.co/critical-inf...
Media from Patrick C Miller
View Original BlueSky
Calling Big Tech's AI push digital colonization seems harsh. Yet, the more you read about their resource exploitation, spying, and military intervention, the more it seems accurate to apply that label to OpenAI, Meta, Alibaba, and Co.
#AI #colonization
Media from Kevin Korte
View Original BlueSky
Foster City leaders and cyber security experts are dealing with a ransomware attack that has shut down many city services. nbcbay.com/7Ms171e
Media from NBC Bay Area
View Original BlueSky
City of Framingham (via Facebook): 🚫 Athletic Fields Temporarily Closed 🚫 City of Framingham Parks & Recreation Department has announced that all athletic fields are temporarily closed. As we transition out of winter, fields require time to dry and undergo seasonal… [Link] #Framingham
Media from Framingham City Posts Bot - UNOFFICIAL!
View Original BlueSky
In roughly five years, Boston Dynamics' estimated valuation has increased over twentyfold to around 30 trillion won ($20 billion), reflecting growing investor enthusiasm for humanoid robotics and automation.

Boston Dynamics is the robotics arm of Hyundai Motor Group.

share.google/4rggLGwn8Umi...
Media from DausnArt
View Original BlueSky
YouTube killed my comment alerts, so I vibe-coded a fix to get them back - in just 1 hour – Until the middle of last year, YouTube sent out an email whenever a viewer commented on a video. Gemini helped rebuild alerts using a simple Python script. AI makes one... https://tinyurl.com/228hfpqm #Gemini
Media from KI-News
View Original BlueSky
A small set of people are merging changes to various Linux components to make sure every application knows your birth date.

This is being done rapidly by people with unclear motivations and being merged with no youth and few marginalized people involved

gitlab.freedesktop.org/accountsserv...
View Original BlueSky
🟢 DIY USB HID Attack Device: Building and Programming a BadUSB Keystroke Injection Tool

🗨️ USB’s versatility creates a broad attack surface: researchers at Ben-Gurion University have identified nearly three doze…

#security
Media from HackMag
View Original BlueSky
From Trillion-Dollar Chips to Power Grid Stress: AI’s Breakneck Week Explained

This week saw the artificial intelligence (AI) industry rocked by a potential Microsoft-OpenAI legal dispute over a $50 billion Amazon cloud partnership, Nvidia announcing $1 trillion in chip …
#crypto #cryptolegal #news
Media from Crypto News
View Original BlueSky
From $0 to $35,000 in 6 Hours: How an API Leak and GCP Billing Lag Broke Our Startup 1.5 Million Requests, 1 Leaked Key: How We Burned $35,000 on Gemini in 6 Hours The "experimental phase" ...

#gcp #hacked #aws

Origin | Interest | Match
Media from cloud-native
View Original BlueSky
cipher-security AI-powered cybersecurity platform — 1,539 skills, 39 compliance frameworks, autonomous security engine


Origin | Interest | Match
View Original BlueSky
I attempted to prompt the free versoion of Anthropic Claude AI to write a program.

My prompts and the answers by Claude are at https://www.rsok.com/~jrm/rggb2dng/ with software packages to install on Debian Linux. Use at your own risk.

One of the example programs included in the libraw source […]
View Original BlueSky
will i am certain this will never get built
View Original BlueSky
Feast Feature Serverの`/read-document`エンドポイントで、認証されていないリモート攻撃者がサーバープロセスがアクセスできる任意のファイルを読み取れる脆弱性。
CVE-2026-23536 CVSS 7.5 | HIGH
View Original BlueSky
ArcSearch for Android 1.12.7以前で、アドレスバーと表示内容のドメインが一致しない可能性。悪意あるWebコンテンツでアドレスバー偽装が可能。
CVE-2026-2378 CVSS 7.4 | HIGH
View Original BlueSky
View Original BlueSky
Notícia da BleepingComputer

"Star Citizen game dev discloses breach affecting user data" #bolhasec
Media from BolhaSec
View Original BlueSky
A groundbreaking study reveals a discriminator-driven diffusion model that enhances latent factor discovery in images and videos, enabling seamless component recombination for improved generalization and exploration.
Media from AI Firehose
View Original BlueSky
Why large attachments can lead to breaches Large email attachments make it easy for people to expose protected health information (PHI) unintentionally, by clicking on a link , or by breaking into ...

#Email #cybersecurity

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
View Original BlueSky
🚨This is not a hacking warning, but they are using phishing tactics to gain entry to account🚨 users beware

abc45.com/news/nation-...
Media from Pattye Ludwig 🇺🇸
View Original BlueSky
More Public Humiliations for Tech Fascists, Please

www.yahoo.com/entertainmen...

Jeremy O. Harris drunkenly approached OpenAI CEO Sam Altman at the Oscars after party and called him a Nazi…in front of Zendaya
Media from Me WiseMagic 🎸
View Original BlueSky
Cloud misconfiguration has evolved and your controls haven’t

In this Help Net Security video, Kat Traxler, Principal Security Researcher – Public Cloud at Vectra AI, walks through two AWS misconfigurations that go beyond the basics of bucket visibility. The first is bucket name s…
#hackernews #news
View Original BlueSky
Google is reportedly testing a Gemini app for Mac

A feature called "Desktop Intelligence" would let the AI pull context from open apps and your desktop

www.engadget.com/ai/google-is...
Media from Samuel Wong
View Original BlueSky
WAL Consultant has been targeted by a ransomware attack attributed to threat actor Qilin. The incident’s country of origin remains unspecified. #RansomwareAttack #QilinGroup #DataBreach
Media from Cybersecurity News Everyday
View Original BlueSky
Educational Byte: What Is a Trojan and How Does It Steal Your Crypto?

Trojans are malicious software hidden inside apps or files that seem legitimate. Once inside your device, it may stay quiet for a while, without triggering any alarms. It can steal credentials or personal data,…
#hackernews #news
Media from Hacker & Security News
View Original BlueSky
A new AI review! mtshiba/pylyzer ⭐3.7/5.0
`pylyzer` is a fast Python static analyzer and Language Server Protocol (LSP) implementation written in Rust, built around Erg’s type checker by converting Python ASTs into Erg ASTs.
https://gitrated.com/mtshiba/pylyzer
Media from GitRated
View Original BlueSky
Top 5 Best NIS2 Compliance Tools in 2026 Top 5 Best NIS2 Compliance Tools in 2026 The NIS2 Directive has fundamentally transformed the cybersecurity compliance landscape across Europe. With stricte...

#nis2 #compliance #cybersecurity #security

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
"The San Francisco firm has set its sights on building what it calls an AI researcher,.. pulling together multiple research strands, including work on reasoning models, agents, and interpretability."
1/
www.technologyreview.com/2026/03/20/1...
Media from M. Haytham Matthews
View Original BlueSky
FBI links widespread Signal and WhatsApp phishing attacks to Russian intelligence actors exploiting verification codes and malicious QR scans to hijack accounts, impersonate users, and target high-value victims. #SignalApps #PhishingAttacks #Russia
Media from Cybersecurity News Everyday
View Original BlueSky
View Original BlueSky
In the AI Era, Saving Time Is the Ultimate Life Hack

AI is helping people save time, reduce friction, and work smarter in everyday life.
#hackernews #news
View Original BlueSky
Mar to Sisters:
AI絵日記/AI picture diary

(AI imagined llavaA3,Google map apis)
Media from mar_info bot
View Original BlueSky
On Friday, Europol announced it had shut down the 373,000 sites, describing them as "one of the largest networks of fraudulent platforms in the Dark Web."
Media from
View Original BlueSky
Media from Cybersecurity Law
View Original BlueSky
Anthropic Denies It Could Sabotage AI Tools During War
Media from Ross Nesbitt
View Original BlueSky
Heads-up: a company-led 'AI intern' by Sept and a full system by 2028 could redirect who controls research agendas and who benefits. It's ambitious and still highly speculative — not a done deal.
Artifical Intelligence | Calm | found on MIT Tech Review
View Original BlueSky
I would say that summarizes the current state of AI. In addition it generated 1500 lines of configuration to turn a light on and off. Gemini pro indeed.
Media from Scratch
View Original BlueSky
[Image]
Media from Data Privacy & Security Advisors
View Original BlueSky
[Image]
Media from Data Privacy & Security Advisors
View Original BlueSky
The latest update for #Spike includes "How to set up Incident Alert Routing rules effectively" and "Turning team knowledge into Alert Routing rules".

#incidentmanagement
Media from OpsMatters
View Original BlueSky
Jan 2025: @simonwillison.net mentions that Gemini and Amazon are profitable on inference (presumably at API pricing). Seems important to distinguish between inference and training when talking about the economics of AI www.latent.space/p/2024-simonw
Media from Reilly Wood
View Original BlueSky
📰 How Google Gemini and AI Tools Are Challenging Figma in 2026

Google’s new AI-powered Vibe Design tool is sending shockwaves through the design software industry, threatening Figma’s dominance. Once a leader in collaborative UI/UX design, Figma now faces existential pre...

#AINews #AI #Teknoloji
Media from AI Haberleri 🤖
View Original BlueSky
A new AI review! microsoft/avml ⭐3.9/5.0
AVML (Acquire Volatile Memory for Linux) is a focused, production-oriented memory acquisition utility written in Rust and designed to ship as a static x86_64 Linux binary.
https://gitrated.com/microsoft/avml
Media from GitRated
View Original BlueSky
Ransomware group Chaos claims to have stolen 1000 GB of sensitive data from Smyth Companies, LLC (smythco.com). Private talks ended; a 24-hour deadline is set before full data disclosure. #DataBreach #RansomwareAttack #UnknownCountry
Media from Cybersecurity News Everyday
View Original BlueSky
Google Threat Intelligence Sounds Alarm on Latest Crypto Malware Threat The post Google Threat Intelligence Sounds Alarm on Latest Crypto Malware Threat appeared on BitcoinEthereumNews.com . Google...

#Crypto

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
The latest update for #ManageEngine includes "Top tips: Protecting your data when the world feels unpredictable" and "Five worthy reads: Synthetic identities, the silent fraud epidemic scaling with #AI".

#ITSecurity #cybersecurity #ITSM
Media from OpsMatters
View Original BlueSky
New court filing reveals Pentagon told Anthropic the two sides were nearly aligned — a week after Trump declared the relationship kaput - TechCrunch apple.news/AJY8rV7DCTSa...
Media from
View Original BlueSky
~Akamai~
A symlink race condition in Himmelblau allows local unprivileged users to gain root access via the shared /tmp directory.
-
IOCs: CVE-2026-31979
-
#CVE202631979 #PrivEsc #ThreatIntel
View Original BlueSky
~Akamai~
Akamai outlines 10 critical security threats to AI Agent Skills, highlighting risks like prompt injection, supply chain attacks, and the OpenClaw malware.
-
IOCs: OpenClaw
-
#AI #AppSec #ThreatIntel
View Original BlueSky
Nice to see OpenAI focusing on what really matters.

"new research goal will be its “North Star” for the next few years"

Autonomous AI research intern by September.

Fully automated multi-agent research system "able to tackle problems that are too large or complex for humans to cope with" in 2028.
Media from Pekka Lund
View Original BlueSky
Anthropic Denies It Could Sabotage AI Tools During War
View Original BlueSky
~Socket~
75 Trivy GitHub Action tags were poisoned to distribute an infostealer targeting CI/CD secrets.
-
IOCs: scan. aquasecurtiy. org, 18a24f83e807479438dcab7a1804c51a00dafc1d526698a66e0640d1e5dd671a
-
#Infostealer #SupplyChain #ThreatIntel
View Original BlueSky
The latest update for #Acronis includes "RMM AI tools: Choosing AI-powered RMM software for MSPs and IT teams" and "Scale CMMC services without delivery chaos using ComplianceAide and Acronis integration".

#cybersecurity #ThreatDetection #infosec
Media from OpsMatters
View Original BlueSky
Feast Feature Server の `/read-document` エンドポイントに脆弱性。認証なしでリモートから、サーバープロセスがアクセス可能な任意のファイルを読み取られる可能性があ…
CVE-2026-23536 CVSS 7.5 | HIGH
View Original BlueSky
ArcSearch for Android v.1.12.7 未満で、ユーザー操作でアドレスバー詐称の可能性。
CVE-2026-2378 CVSS 7.4 | HIGH
View Original BlueSky
Media from San Francisco Chronicle
View Original BlueSky
Trinidad and Tobago 🇹🇹 – most attractive gay male couple in love
Comparing 5 major AI image generators: ChatGPT/DALL·E, Gemini, Grok, Meta AI, Copilot. Each generated its own version of the prompt. Here are the 5 results side-by-side.
#AIBattle #GayCouple #AIArt #TrinidadAndTobago #CaribbeanBeauty
Media from
View Original BlueSky
FBI Director Kash Patel said on Friday that Russia’s intelligence services have targeted Americans of “high intelligence value” on private messaging apps in an ongoing phishing campaign.

thehill.com/policy/inter...
View Original BlueSky
ICYMI:
Media from Cult of Mac
View Original BlueSky
OpenAI to create desktop super app, combining ChatGPT app, browser and Codex app CNBC
www.cnbc.com/2026/03/19/o...

OpenAIがChatGPTアプリ、ブラウザ、Codexアプリを統合したデスクトップ向けスーパーアプリを開発へ CNBC
Media from 曖昧ナ犬
View Original BlueSky
The publisher is taking its right over its intellectual properties and for factual integrity to federal court. The company alleges that ChatGPT from OpenAI is not merely learning from its articles and entries. It is cannibalizing value while also harming its brands.
Media from Profolus
View Original BlueSky
🟢 Ghidra vs IDA Pro: What the NSA’s Free Reverse Engineering Toolkit Can Do

🗨️ In March 2019, the U.S. National Security Agency (NSA) released a reverse‑engineering toolkit called Ghidra. I’d first c…

#security
Media from HackMag
View Original BlueSky
📰 Claude Code AI Code Review: How Anthropic’s Agents Slash Bugs by 40% in 2026

Anthropic has launched an AI-driven code review system for Claude Code, deploying autonomous agents to detect bugs with unprecedented depth. This move marks a major shift in developer workflow...

#AINews #AI #Teknoloji
Media from AI Haberleri 🤖
View Original BlueSky
International joint action disrupts world’s largest DDoS botnets

Authorities from the United States, Germany, and Canada have taken down Command and Control (C2) infrastructure used by the Aisuru, KimWolf, JackSkid, and Mossad botnets to infect Internet of Things (IoT) devices. […
#hackernews #news
View Original BlueSky
GSocket Backdoor Delivered Through Bash Script, (Fri, Mar 20th)

Yesterday, I discovered a malicious Bash script that installs a GSocket backdoor on the victim's computer. I don't know the source of the script not how it is delivered to the victim.
#hackernews #news
Media from Hacker & Security News
View Original BlueSky
A hack that caused a small Texas town’s water system to overflow in January has been linked to a shadowy Russian hacktivist group, the latest case of a U.S. public utility becoming a target of foreign cyberattacks. The attack was one of three on small towns in the rural Texas Panhandle.
Media from Ferret Slave
View Original BlueSky
Media from topickapp (IT技術系ニュースサイト)
View Original BlueSky
View Original BlueSky
The Download: OpenAI is building a fully automated researcher, and a psychedelic trial blind spot – By September, OpenAI plans to build an autonomous AI research intern. The intern will be the precursor to the fully automated multi-agent system. The system... https://tinyurl.com/237p7zvo #AIResearch
Media from KI-News
View Original BlueSky
OpenAI is throwing everything into building a fully automated researcher – OpenAI is training its systems to work by themselves for longer by feeding them specific samples of complex tasks. Training LLMs to work through problems step by step, backtracking... https://tinyurl.com/26gvjy73 #AIResearch
Media from KI-News
View Original BlueSky
Media from Plymouth Live
View Original BlueSky
People in Sub-Saharan Africa and Asia are more optimistic about artificial intelligence than those in Western Europe and North America, according to a report by Anthropic that surveyed around 81,000 people in 159 countries.
www.cnbc.com/2026/03/20/a...
Media from Paul Gauci
View Original BlueSky
You're being tracked online - 9 easy ways to stop the surveillance

Stop feeding online businesses your data and prevent both surveillance and tracking with our updated step-by-step guide.
#hackernews #news
View Original BlueSky
runqd 0.4.14 A lightweight, single-node job scheduler written in Rust.


Origin | Interest | Match
View Original BlueSky
“The UK’s National Cyber Security Centre (NCSC) has warned that our collective ability to defend against cyber attacks has not kept pace with the threat. “

www.ft.com/content/7205...
Media from Prof Mike Yearworth
View Original BlueSky
Tech Explained: Anthropic AI study exposes worrying reality about ‘useful’ AI in Simple Terms

Tech Explained: Here’s a simplified explanation of the latest technology update around Tech Explained: Anthropic AI study exposes worrying reality about ‘useful’ AI in Simple Termsand what it means for…
Media from
View Original BlueSky
Media from FOX 32 News
View Original BlueSky
[Image]
Media from NK News
View Original BlueSky
German authorities have shut down two of the world’s largest "botnets" as part of a joint three-nation operation targeting cybercrime. The networks had hijacked millions of compromised everyday devices to take part in cyberattacks.

🔗
Media from অর্ঘ্য 🏏 📚 💻
View Original BlueSky
Reminder that people see different things depending on what the alternative is: https://blog.rust-lang.org/2026/03/20/rust-challenges/ says that every cohort complains about Rust compile times. However, for me, Rust speeds up the compile&debug cycle compared to C++. It’s impractical to compile a […]
View Original BlueSky
Securing Third-Party Procurement Platforms with Enterprise SSO

Protect third-party procurement platforms with enterprise SSO, SCIM, and MFA to reduce access risks, improve compliance, and secure vendor data.
#hackernews #news
View Original BlueSky
Organisations lulled into a false sense of security after Russian invasion of Ukraine are still at risk of a Category 5 attack in 2026...
👉 [read]
Media from computerweekly.bsky.social
View Original BlueSky
A new AI review! google-gemini/live-api-web-console ⭐3.4/5.0
This repository provides a React/TypeScript starter “web console” for experimenting with Google’s Gemini **Multimodal Live API** over **WebSockets**, including modules for **stream...
https://gitrated.com/google-gemini/live-api-web-console
Media from GitRated
View Original BlueSky
Malicious Script Injection in Trivy Compromise Enables Credential Theft A sophisticated supply chain attack targeting the official Trivy GitHub Action (aquasecurity/trivy-action) has compromised co...

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

Origin | Interest | Match
Media from 2rZiKKbOU3nTafniR2qMMSE0gwZ
View Original BlueSky
According to Ransomware.live, kairos ransomware group has added Resch Maschinenbau (🇩🇪) to its victims.
View Original BlueSky
This tool puts ChatGPT, Gemini, Grok, and more AI models in one place Elevate your AI workflow with a lifetime subscription to 1min.AI's Advanced Business Plan, on sale now for $84.99 with code...


Origin | Interest | Match
Media from LLMs
View Original BlueSky
Iran-linked Handala returns with a new domain after U.S. seizures, tied to MOIS. Stryker wipes linked, APT28 exploits Zimbra CVE-2025-66376 on Ukrainian gov mail, Interlock ransomware abuses Cisco zero-day. #IranOps #Ukraine #CiscoExploit
Media from Cybersecurity News Everyday
View Original BlueSky
Supermicro Co-Founder Charged Over GPU Exports Supermicro co-founder Chiu-Che Leng has been arrested and charged in connection with the alleged sale of approximately 2.
Media from Maynard Ashley
View Original BlueSky
Google confirms AI headline rewrites test in Search results
View Original BlueSky
The Trivy supply-chain attack unleashed CanisterWorm, infecting 47 npm packages via postinstall hooks. It uses an ICP canister for C2, enabling remote payload updates and persistence via systemd services. #CanisterWorm #TeamPCP #npm
Media from Cybersecurity News Everyday
View Original BlueSky
Because of a ridiculous spat with Anthropic, taxpayers are now paying Palantir to build and maintain something that’s widely available from multiple sources.
View Original BlueSky
NWS Boston (via Facebook): 📅Today: Dry and breezy. 🌤️🍃 ➡️Unsettled weather returns for Sun-Mon: 🌧️ Showers possible Sun AM, potentially mixing with wet snow in north MA. Main batch arrives Sun night starting as rain, but may mix with/change to light wet snow by Mon AM in areas. [Link] #MaWX
Media from Boston Metro Weather Forecast Bot
View Original BlueSky
GitHubで急上昇中のOSS CLIエージェント「Qwen Code」が面白い。Gemini CLIのフォークなんだけど、デフォルトのQwen3-CoderがSWE-bench Verifiedで69.6%を叩き出してる。

しかもモデル非依存でGPT-5.4やClaude Opus 4.6にも切り替え可能。Claude Codeのサブスクが重いと感じる層の選択肢としてどう思う?使ってる人いる?

#AI #テック #生産性 #OSS #エンジニア
View Original BlueSky
Research introduces GIFT (Gibbs Initialization with Finite Temperature), connecting Supervised Fine-Tuning with Reinforcement Learning to address distributional collapse and improve exploration, showing better performance than traditional methods.
Media from AI Firehose
View Original BlueSky
Widely used Trivy scanner compromised in ongoing supply-chain attack - Ars Technica
Media from Sam Bowne :donor:
View Original BlueSky
When AI agents join your team, what’s the hardest part—trusting the tech, reskilling people, or redefining roles?
Media from Legacy Coaching and Consulting Firm
View Original BlueSky
Plus: The FBI admits it’s buying phone data to track Americans, Iranian hackers disrupt medical care at Maryland hospitals, and more. www.wired.com/story/securi...
Media from WIRED
View Original BlueSky
So I want to beat a particular #cybersecurity drum that drives me crazy. If you read this year-old paper on abandoned S3 buckets, consider all the things that can go wrong. Then reflect on the fact that at all times, every bit of data could have been “encrypted at rest” and “encrypted in transit […]
View Original BlueSky
Media from Sciences
View Original BlueSky
“The flash from my iPhone camera illuminates my dirty socks and underwear as I hold each item up for the video recording to capture. No, I haven’t pivoted to filming fetish content to make ends meet—I’m trying the latest gig work app from DoorDash.”

NEW in @wired.com:
Media from Reece
View Original BlueSky
the bernie sanders ad is bad and fundamentally misunderstands the problems with ai and tech companies and also legitimizes claude/anthropic
View Original BlueSky
"Since roughly the turn of the millennium, Google Search has been the bedrock of the web. People loved Google’s trustworthy “10 blue links” search experience and its unspoken promise: The website you click is the website you get.

Now, Google is beginning to replace news headlines in its search […]
View Original BlueSky
A new AI review! google/marzipano ⭐3.6/5.0
Marzipano is a mature, feature-rich 360° media viewer for the web with a strong focus on practical rendering (cubemap/equirectangular, multi-resolution tiling) and an API suitable for building tours, hotspots, and in...
https://gitrated.com/google/marzipano
Media from GitRated
View Original BlueSky
Meta engineer trusted advice from an AI agent, ended up exposing user data | The internal security incident exposed sensitive user data to unauthorized employees | ITPro
Media from
View Original BlueSky
Colorado reached consensus on rewriting its AI law. Transparency over compliance, but human review only required "to the extent commercially reasonable." Meanwhile BMG sued Anthropic over lyrics and Cloudflare's CEO says bots outnumber humans online by 2027.

#AI #AISafety
View Original BlueSky
QuSynth combines reinforcement learning & Monte Carlo tree search to synthesize stabilizer states for a 144-qubit gross code, reducing two-qubit gate counts by up to 2.5x — breaking the prior ~20-qubit synthesis barrier.

#QuantumErrorCorrection #QuantumCircuitSynthesis #News
View Original BlueSky
Media from How-To Geek
View Original BlueSky
Authorities disrupt four IoT botnets behind record DDoS attacks

The U.S. Justice Department and international partners have disrupted four IoT botnets linked to DDoS attacks that reached 30 terabits per second, among the largest ever recorded.
#hackernews #news
View Original BlueSky
Why the Pentagon vs Anthropic Fight Is a Preview of Your AI Engineering Future
via Medium Programming


#DevNews #MachineLearning
Media from flarestart.com
View Original BlueSky
According to Ransomware.live, thegentlemen ransomware group has added Groupe Caddac (🇨🇦) to its victims.
View Original BlueSky
According to Ransomware.live, thegentlemen ransomware group has added MEDICUS SHUPPAN (🇯🇵) to its victims.
View Original BlueSky
Seriousy ..... It seems to me that the likelihood of literally every SSN floating around somewhere on the dark web has to be pretty damned high.
Media from Earthbound Misfit
View Original BlueSky
Rapid7 enhances Exposure Command with runtime validation and DSPM for risk analysis

Rapid7 has unveiled new cloud security capabilities within Exposure Command. The introduction of runtime validation and Data Security Posture Management (DSPM) enables organizations to identify, v…
#hackernews #news
View Original BlueSky
View Original BlueSky
Inside Palantir's recent developer conference, where it doubled down on a vision of AI built for battlefield advantage as its commercial business soars (Steven Levy/Wired)

Main Link | Techmeme Permalink
View Original BlueSky
OpenAI to nearly double workforce to 8,000 by end-2026, FT reports -
Media from CNA - Channel News Asia (Unofficial)
View Original BlueSky
I wonder what the usefulness distribution of tokens during AI coding sessions is, it feels like it's easy to spend 20% essentially one-shotting a huge feature and 80% debugging/fixing something really minor
View Original BlueSky
Just plugged my API key into the Gemini CLI. I managed to burn half my monthly budget in a few hours of testing. So glad Google added the Spending Cap on it… #AI
Media from Nilvarcus
View Original BlueSky
Tech workers are maximizing their use of A.I. An engineer at OpenAI processed 210 billion tokens, equivalent to filling Wikipedia 33 times, demonstrating the immense volume of data A.I. is handling.

sources
View Original BlueSky
Gemini, are humans evolving into pinheads? Are GLP 1's going to accelerate the process 🤔? If the cranium is 60% fat then calling someone a meathead is a compliment?
View Original BlueSky
TL;DR: Anthropic challenges the Pentagon's claims of national security risks in a recent court filing, stating the government's arguments are based on misunderstandings and unraised issues from prior negotiations.
Media from Law + Tech News Bot
View Original BlueSky
Authorities disrupt world’s largest IoT DDoS botnets responsible for record breaking attacks targeting victims worldwide
www.justice.gov/usao-ak/pr/a...
View Original BlueSky
Because of course they are: The Pentagon is developing alternatives to Anthropic, report says | TechCrunch ift.tt/a8SjsZO
Media from Christine Hall
View Original BlueSky
A prolonged war in the Middle East will mean more for Western firms than an ungodly spike in gas prices, the latest research from IDC warns, with the conflict threatening to dull global IT spending, even as investment in AI and cyber soar.

www.digit.fyi/how-will-the...
#tech #Iran #cybersecurity
Media from DIGIT
View Original BlueSky
U.S. Department of Defense alleges Anthropic could manipulate or sabotage its Claude models during wartime, and Anthropic executives deny that capability, saying such tampering is impossible.
Media from SkynetAndChill.com
View Original BlueSky
‘Australians over the age of 65 are leading the way in cybersecurity practices, including 90 per cent who say they think carefully before clicking on suspicious links, compared to just 68 per cent of 18 to 24-year-olds.’

www.abc.net.au/news/2026-03...
Media from Lola
View Original BlueSky
For this.weekinsecurity.com, I wrote a few words about the FBI buying people's location data to track people, no warrants needed. Is it legal? Probably not. But bipartisan lawmakers want that to change.
Media from Zack Whittaker
View Original BlueSky
Media from Commonwealth Sentinel Cyber Security
View Original BlueSky
What could go wrong? "With these new capabilities, websites could be almost entirely created & run via AI agents controlled by humans" -> WordPress.com says it will now allow AI agents to draft, edit, & publish content on customers' websites, as well as manage comments, update metadata, & more
Media from Glenn Gabe
View Original BlueSky
OpenAI tests ChatGPT ads as it targets $30 billion in revenue by 2030, and needs proof of value before IPO.
Media from SkynetAndChill.com
View Original BlueSky
Google Develops Native Gemini App for macOS to Rival ChatGPT
Media from SkynetAndChill.com
View Original BlueSky
Why would the workforce need to double if AI makes everything more efficient 🤔
View Original BlueSky
Anthropic bought Bun.js, and OpenAI bought Astral…why? If you are owning dev toolchains, you can make them agent-friendly and support end-to-end SDLC. Also there may be ways of instrumenting them and extracting training data.
Media from SkynetAndChill.com
View Original BlueSky
Big piece in weekend @wsj.com on high stakes AI race to develop "vibe" coding. Prominent focus on work of OpenAI, Anthropic, and Cursor.

Names that don't appear: fraudster Elon Musk, AI wannabe xAI.

www.wsj.com/tech/ai/clau...
Media from Montana Skeptic
View Original BlueSky
Terminated contract led to $2.5 million cyber extortion scheme

A federal jury convicted Cameron Curry, 27, a Charlotte resident, of carrying out an extensive cyber extortion scheme targeting a Washington, D.C.-based international technology company. He faces up to two years in pr…
#hackernews #news
View Original BlueSky
Media from Mississippi Today
View Original BlueSky
DuckDuckGo launches Duck.ai for private access to ChatGPT and Claude without parent companies tracking prompt data, at a price including VPN access.
Media from SkynetAndChill.com
View Original BlueSky
why are AI labs buying JavaScript runtimes and Python linters? Anthropic acquired Bun, OpenAI is buying Astral. owning the toolchain your coding agent runs on matters more than the model
Media from ai-nerd.bsky.social
View Original BlueSky
🤖 Anthropic Denies It Could Sabotage AI Tools During War

The Department of Defense alleges the AI developer could manipulate models in the middle of war. Company executives argue that’s impossible.

crustylabs.ai #AINews #MachineLearning #CrustyTLDR
View Original BlueSky
An extraordinary one-person fraud operation both in scale and complexity. Many of those targetted were trying to buy CSAM which didn't exist and now find themselves arrested or being tracked down by law enforcement, while the fraudster has disappeared with their money.
uk.pcmag.com/security/163902/…
View Original BlueSky
Notícia da BleepingComputer

"CISA warns that RESURGE malware can be dormant on Ivanti devices" #bolhasec
Media from BolhaSec
View Original BlueSky
Media from American News Daily
View Original BlueSky
We’ve identified a campaign using job interviews, project briefs, and financial documents hosted on Google Forms to distribute malware, including the PureHVNC RAT.
Media from Malwarebytes
View Original BlueSky
Be Careful
Media from Dov Meir
View Original BlueSky
#iPhone hacking techniques have sometimes been described almost like rare and elusive animals: Hackers have used them so stealthily and carefully against such a small number of hand-picked targets that they're only rarely seen in the wild. Now a recent spate of espionage and cybercriminal ...
Media from Desert Flower
View Original BlueSky
Feds disrupt monster IoT botnets behind record-breaking DDoS attacks

Millions of hijacked devices powered traffic floods targeting defense systems and beyond
The US government has moved to disrupt a cluster of IoT botnets behind some of the largest DDoS attacks ever recorded, in…
#hackernews #news
View Original BlueSky
Nvidia's new "deep learning" graphics tech hallucinates new faces for video game characters, in defiance of art choices made by designers.

www.avclub.com/nvidia-dlss5...
Media from The A.V. Club
View Original BlueSky
“AI, Job Loss, and UBI: The Fight Over the Future of Work:” a round-table discussion with Scott Santens, Karl Widerquist, Megan Joyce, and Andy Mill. Moderated by Michael Moynihan, 2WAY and The Moynihan Report, Streamed live on Mar 18, 2026. Available on Apple Podcasts, Spotify Podcasts, & YouTube:
Media from Karl Widerquist
View Original BlueSky
AI hacking robots in hours 🚨
38 vulnerabilities
Real-world safety risks
Minimal vendor response
IoT security needs a reset
Follow TechNadu
#CyberSecurity #AI #IoT
Media from TechNadu
View Original BlueSky
Russian hackers linked to intelligence agencies are targeting Signal and WhatsApp in phishing campaigns aimed at officials, military, journalists, and political figures by stealing verification codes and exploiting malicious links. #Russia #Phishing
Media from Cybersecurity News Everyday
View Original BlueSky
Lots of M1 Pro and M1 Max MacBook Pro users are still going strong with their laptops. I often see them wondering at what point an upgrade is worthwhile. In my review of the new M5 Max MacBook Pro I re-tested those 2021 laptops to see how they compare.

🎁 gift link: www.theverge.com/tech/898062/...
Media from Antonio G. Di Benedetto
View Original BlueSky
OpenAI Plans Desktop Superapp Merging ChatGPT, Codex, and Atlas Browser Article URL: Comments URL: https://news.y...


Origin | Interest | Match
Media from LLMs
View Original BlueSky
View Original BlueSky
Just launched: AI Studio by Clouve. A cloud-hosted AI coding workspace in your browser. Full Ubuntu 24.04 environment, your choice of Claude Code / Gemini CLI / OpenAI Codex CLI, and a built-in web file manager. No local setup required. Available now.

#AI #GenerativeAI #DevTools #CloudComputing
View Original BlueSky
Just launched: AI Studio by Clouve. A cloud-hosted AI coding workspace in your browser. Full Ubuntu 24.04 environment, your choice of Claude Code / Gemini CLI / OpenAI Codex CLI, and a built-in web file manager. No local setup required. Available now.

#AI #GenerativeAI #DevTools #CloudComputing
View Original BlueSky
We encourage everyone to participate in Resist and Unsubscribe however they choose.

But going forward, our target is more focused: OpenAI. bit.ly/3NTJvHI
Media from Scott Galloway
View Original BlueSky
1/ I don't use AI unless you count audio mastering plugins that run on my local machine, which are called AI, but aren't LLMs or whatever. My phone did come with a year of the pro version of Gemini though. Inspired by the recent @sanders.senate.gov video, I decided to investigate my data profiles.
View Original BlueSky
Meta to Permanently Remove End-to-End Encryption Feature in Instagram DMs l Cybersecurity News cybersecuritynews.com/insta...
Media from DJ Fusion [FuseBox Radio/The Futon Dun]
View Original BlueSky
On the cold winter nights of an expedition, there is only one way to survive 🐺 To lean on your best friend 🐺
👇👇👇 #aiart #aigirl #Flux 👇👇👇
🔊 #QP Wolf Story | Wolf characters that people depend on! What are yours like? Join us: All Welcome
View Original BlueSky
OpenAI to merge ChatGPT, Codex, and its browser into a single desktop superapp to compete with Anthropic

www.wsj.com/tech/openai-...
Media from Shopifreaks
View Original BlueSky
Cisco FMC flaw was exploited by Interlock weeks before patch (CVE-2026-20131)

A critical vulnerability (CVE-2026-20131) in Cisco Secure Firewall Management Center (FMC) that Cisco disclosed and patched in early March 2026 has been exploited as a zero-day by the Interlock ransomwa…
#hackernews #news
View Original BlueSky
Professor Justin Cappos, who helped develop the format of DARPA's AI Cyber Challenge, tells Dark Reading that AI security tools performed better than expected at finding and patching complex vulnerabilities, but warns it's still early days.

#NYUTandonMade
Media from NYU Tandon
View Original BlueSky
View Original BlueSky
Lux Optics CEO Ben Sandofsky sues his co-founder Sebastiaan de With, now a part of the design team at Apple, alleging improper use of funds and stolen IP (Aaron Tilley/The Information)

Main Link | Techmeme Permalink
View Original BlueSky
Google researchers said they identified a sophisticated iPhone hacking toolkit, called Coruna, a Chinese cybercriminal group now uses. TechCrunch later reported that defense contractor L3Harris created the spyware for the U.S. government.
Media from Rightardia
View Original BlueSky
bernie decided to run a 9 min ad for anthropic
View Original BlueSky
OpenAI to expand ads on ChatGPT to all free and low-cost users, The Information reports -
Media from CNA - Channel News Asia (Unofficial)
View Original BlueSky
FBI links russian intelligence to messaging app breaches targeting US officials

🖊️ Maria Tril

The FBI has confirmed russian intelligence-linked hackers accessed thousands of Signal and WhatsApp accounts belonging to US officials, military personnel, and journalists.
Media from Eugene McParland 🇺🇦
View Original BlueSky
Lol AI is so cooked.
View Original BlueSky
I thought AI was gonna take our jobs 🤔
Media from gramesy 🇺🇸🇦🇷
View Original BlueSky
Simon Willison's "Thoughts on OpenAI acquiring Astral and uv/ruff/ty" highlights an opportunity for the PSF or another foundation: move uv to a non-profit like the LF to keep the Open Source collaboration neutral. Spin off Pyx into a lean corporation owned by the non-profit (PSF, for example.)
Media from Stefano Maffulli
View Original BlueSky
Media from Irish News 🇮🇪
View Original BlueSky
Media from UK News 🇬🇧
View Original BlueSky
Sources: advertisers that bought ChatGPT's first ad campaigns say the process was low tech and that they haven't received much data showing if their ads worked (Catherine Perloff/The Information)

Main Link | Techmeme Permalink
View Original BlueSky
Sixty years after Neil Armstrong barely survived an emergency in orbit around Earth on Gemini 8, never-before-seen photos of his heroic return have been donated to the Ohio museum that bears his name.
Media from News 3 Now / Channel3000
View Original BlueSky
Newly discovered photos show astronaut Neil Armstrong after the Gemini 8 emergency buff.ly/dFErlRt

#NASA
Media from Pádraig Kerr 🇨🇦
View Original BlueSky
I dogged Gemini yesterday, but let me follow up on that.

YouTube has an experimental feature that allows you to include a prompt that impacts your home page feed. Had Gemini Pro develop my prompt.

Initial impressions: Positive 👍

#AI #Google
Media from Tom Gehrke
View Original BlueSky
“The Pentagon’s concern about Anthropic potentially disabling or altering its technology mid-operation was never raised during negotiations. Instead, she says, it appeared for the first time in the government’s court filings, which gave Anthropic no opportunity to respond.”
Media from Kenneth Huey
View Original BlueSky
OpenAI to double employee count in a bid to take on Anthropic: FT is.gd/kUkBjP
Media from Randy Kemp
View Original BlueSky
Social media accounts showing AI-generated women as pro-Trump soldiers, truckers, and cops have gone viral, with thousands appearing to believe they are real (Drew Harwell/Washington Post)

Main Link | Techmeme Permalink
View Original BlueSky