Mania Africa 💎🌍

City of Framingham (via Facebook): Live in District 8? The Framingham Parks & Recreation Commission will have a discussion about amenities in that district as part of its monthly meeting TONIGHT, April 14 at 7 p.m. Agenda, background materials, & Zoom link - https://framinghamma.… [Link] #Framingham

Google Gemini 3.1 Pro is here to solve the "unsolvable." With double the reasoning performance of its predecessor, it’s designed for the most complex enterprise challenges. Is your dev team ready?

https://www.infoworld.com/article/4134809/google-gemini-3-1-pro-boosts-complex-problem-solving.html

New Claude Desktop app for MacOS loving the task and preview panes. Demo of me using my Claude ai-image-creator skill to edit an existing image with the screenshot of new Claude Desktop using Google Nano Banana 2 for image editing within Claude Code for Desktop 😎 ai.georgeliu.com/p/building-a...

Google introduces “Skills” in Chrome to make Gemini prompts instantly reusable: Google has introduced a new feature called "Skills" in its Chrome browser, designed to enhance the usability of its AI chatbot, Gemini. This innovation allows users to save frequently used… https://ranked.news/645750?u=b

Spacebears Launches Ransomware Attack on Ultimate Metals Summary Introduction to Malware Binary Triage (IMBT) Course Looking to level up your skills? Get 10% off using coupon code: MWNEWS10 for any...

#Malware #News

Origin | Interest | Match

Banks Are Warned About Anthropic’s New, Powerful A.I. Technology www.nytimes.com/2026/04/10/b...

Crypto investors be advised you will be SOL when AI steals your bitcoin.

www.bleepingcomputer.com/news/securit...

🤖 今朝のAIニュース

1/ Anthropicの共同創業者がトランプ政権にMythosについてブリーフィングしたことを確認
2/ Max HodakのScience Corpが人間の脳に初めてセンサーを埋め込む準備を進めている
3/ GoogleがChromeにAIスキルを追加し、お気に入りのワークフローを保存できるように

#AI #テック

Anthropic: A nearly $1,000,000,000 cyber security threat.

Anthropic kills enterprise flat fees, forces per-token billing. Even Retool CEO who liked Claude better switched to OpenAI over uptime. 98.95% vs 99.99% standard. #AI https://www.implicator.ai/anthropic-shifts-enterprise-billing-to-per-token-pricing-the-flat-fee-era-is-over/

Google is expanding Personal Intelligence to Gemini users globally and it’s a huge shift #Technology #SoftwareandApps #Other #Google #Gemini #PersonalIntelligence

Trusted access for the next era of cyber defense (https://news.ycombinator.com/item?id=47770770)

icelandmonitor.mbl.is/news/politic...

There are a very large number of security fixes in Windows today. infosec.exchange/@briankrebs/...

CFPsec v202: The Ultimate Cybersecurity Conference Tool That Automates Your CFP Research + Video

Introduction: Call for Papers (CFP) deadlines for top cybersecurity conferences are notoriously hard to track, often causing researchers to miss submission windows. CFPsec v2.0.2 is a Python-based…

In the Wake of Anthropic’s Mythos, OpenAI Has a New Cybersecurity Model—and Strategy www.wired.com/story/in-the...

Novo Nordisk partners with OpenAI in weight loss drug race thehill.com/policy/healt...

Developers and power users are accusing Anthropic of quietly degrading Claude’s performance, which Anthropic denies

venturebeat.com/technology/i...

These idiots from BlueOwl and Goldman Sachs started this nightmare why are we not going after them using Anthropic

A study published in Digital Medicine concluded that LLM from Meta, ChatGPT, Anthropic, and Google have “the potential to lead to serious patient harm.” As it happens, Meta, Amazon, and Google have all benefited from tax cuts that were partially offset by reductions in healthcare spending.

#Computers #and #the #Internet #Artificial #Intelligence #Computer #Security #ChatGPT #Anthropic #AI

Origin | Interest | Match

"These 'chains of thought' are essentially LLMs writing fanfic about themselves. Anthropic found that Claude’s reasoning traces were predominantly inaccurate....imagine a gang of children shouting out make-believe computer phrases while watching the washing machine run." aphyr.com/posts/411-th...

I built an Umbraco.AI plugin to evaluate pages for things like content quality, brand alignment, SEO, GEO, or basically anything you want to prompt! Check it out and let me know what you think!

github.com/ProWorksCorp...

What are your thoughts? #umbraco #ai

💻 Anthropic's Claude Opus 4.6 upgrades coding, agentic tasks, and planning with 1M token context and 128k outputs, challenging OpenAI as their smartest model yet.[1][2][3]

📰 NDTV
🔗 https://www.ndtv.com/world-news/anthropic-unveils-latest-ai-model-as-openai-rivalry-intensifies-10954933

#AI #T...

OpenAI CEO Sam Altman’s home was attacked twice in three days, marking a sharp escalation in anti-AI sentiment.

Claude Code routines let AI fix bugs and review code on autopilot

Anthropic has introduced "routines" for Claude Code that can automatically fix bugs, review pull requests, and respond to code events without requiring a user's local machine. These automated processes allow…

Via The Decoder

Don't miss Ryan D’Cunha's comprehensive guide to GPUs, where you'll learn how to optimize GPU efficiency through understanding architecture, bottlenecks, and fixes ranging from simple PyTorch commands to custom kernels.

Are Silos Preventing Your Back Office and Field Teams From Collaborating?

If you’re constantly experiencing project delays and putting out data fires, your business could be dealing with an ongoing silo.
#hackernews #news

⚡ Gemini app starts rolling out Personal Intelligence globally

#BigTech #Tech

New post: Anthropic Shares AI Risks With U.S. Government https://bohiney.com/anthropic-shares-ai-risks-with-u-s-government/?utm_campaign=bluesky&utm_medium=twitter&utm_source=twitter

I replaced my entire IT department with 14 AI agents. Week 1 results: 💥🚀💻⏳
Get YOUR own AI-driven cloud infrastructure!
NovaOS lets you harness cutting-edge

https://novaos.cloud
#AIAgents and more, without compromising on security or privacy.
Secure...
#TechStartup #DeepLearning #PrivateAI...

Texas man attacks OpenAI CEO home with Molotov! AI fear sparks violence #AI #Tech #Safety #FBI #Molotov #Texas

OpenAI Expands Cybersecurity Program Before Deploying New Models OpenAI plans to expand access to a program introduced in February that gives cybersecurity professionals access to frontier models. ...

#artificial #intelligence #AI #Cybersecurity #GPT-Cyber #News […]

[Original post on pymnts.com]

Flash Charm Inc, linked to US tech firm Idera, faces a ransomware attack by threat actor coinbasecartel. Incident highlights risks in enterprise IT software supply chains. #UnitedStates #EnterpriseIT #RansomwareAttack

The man accused of trying to kill OpenAI CEO Sam Altman by throwing a Molotov cocktail at his San Francisco home was experiencing a mental health crisis and has been overcharged by prosecutors, his public defender said Tuesday.

phpの依存管理ツールcomposerの脆弱性が修正された。composer.phar self-update でアップデート推奨 "Composer 2.9.6: Perforce Driver Command Injection Vulnerabilities (CVE-2026-40261, CVE-2026-40176)" https://blog.packagist.com/composer-2-9-6-perforce-driver-command-injection-vulnerabilities/

OpenAI and Anthropic Are Racing to Build AI Cyber Weapons. Neither Will Let You Use Them. #automation #future #cybersecurity

🚨 Have you or someone you know been impacted by a #scam? You're not alone. 73% of US adults have been affected by scams in the last year, leading to over $16 billion reported lost. Imposter scams alone cost Americans $3 billion.

[ www.axios.com/2025/04/23/fbi-internet-crime-loss-record-high-2024 ]

👀

📍 Singapore bound for #BHA2026!
Stop by DomainTools booth # 119 to see how our integrations reduce context-switching and identify evolving threats in real-time.
Don't leave your 2026 strategy to chance. Schedule a chat➡️
#BlackHatAsia #CyberSecurity

JPMorgan Chase CEO Jamie Dimon said Tuesday that while artificial intelligence tools could eventually help companies defend themselves from cyberattacks, they are first making them more vulnerable.

Read more from Dimon: cnb.cx/4tKCd8m

💻 Apple threatened to yank Grok from App Store over viral sexualized deepfakes of women & minors, per NBC letter to senators. Advocacy groups now demand Apple/Google boot Grok & X apps entirely.[1][4...

📰 9to5Mac
🔗 https://9to5mac.com/2026/04/14/apple-reportedly-threatened-to-remove-grok-from...

Netgear Gets First Conditional Approval for Router Sales in the US

In a letter to senators sent in January, Apple detailed how it responded to Grok’s ability to generate sexualized deepfakes. #elon #musk #elonmusk
https://mu.sk/HZtD

OpenAI Engineer Helps Companies Boost Sales – Sarang Gupta is a member of the data science staff at OpenAI in San Francisco. He works with the go-to-market team to help businesses adopt ChatGPT and other products. Gupta says he tries to ensure his work has... https://tinyurl.com/2yoel3jy #AIPowered

sn-news: #security #comms #industry Cyber-attacks could disrupt smart factories by targeting time itself techxplore.com/news/2026-02...

www.bloomberg.com/news/article...

Cybersecurity jobs available right now: April 14, 2026

Cyber Security Engineer/Application Security Specialist Tecnots | India | On-site – View job details As a Cyber Security Engineer/Application Security Specialist, you will integrate security into the SDLC, perform …
#hackernews #microsoft #news

💻 Meta's AI leaders at Superintelligence Labs are discussing partnerships to integrate Google Gemini and OpenAI models into Meta AI apps as interim boosts.[1][2][5]

📰 The Information
🔗 https://www.theinformation.com/articles/metas-ai-leaders-discuss-using-google-openai-models-apps

#AI #Tech ...

The inaccuracies of Google's #AI overviews is creating a crisis of misinformation: futurism.com/artificial-i... #ArtificialIntelligence

[Image]

Reporting like this makes me want to scream because "[AI models] solve cybersecurity problems 93% of the time, up from 15% in 2024" is 1) not true, and 2) not even necessarily a good thing.

www.forbes.com/sites/steven...

A cyberattack using 'wiper' malware hit Poland's energy grid, exposing critical infrastructure vulnerabilities. As society grows more digital, strengthening energy resilience is increasingly essential.

Source

Kraken faces extortion from hackers threatening to release videos of internal client-support systems after insider breaches involving support employees accessing limited customer data. No funds at risk, law enforcement involved. #Kraken #InsiderThreat

All of the AI labs ought to have a roon who's willing and able to publicly disagree with the company when warranted. But would the leaders of Anthropic/GDM tolerate such public disagreement?

Network Security Exposed: The CIA Triad, Zero Trust, and How to Stop DDoS & BGP Hijacking – A Technical Deep Dive + Video

Introduction: Network security is no longer just about firewalls and VPNs—it is a layered defense strategy built on the CIA triad: Confidentiality, Integrity, and Availability.…

WindowsのRDPファイル悪用対策。フィッシング攻撃に対し、警告表示や共有リソース無効化で保護強化。

www.wired.com/story/how-th...

Hybrid Hierarchical Federated Learning over 5G/NextG Wireless Networking
tech_blogs_arxiv | Author: Haiyun Liu, Jiahao Xue, Jie Xu, Yao Liu, Zhuo Lu

Krakenは、内部情報漏洩により、顧客データを持つ内部システム動画の公開を脅迫された。

We’re Using So Much AI That Computing Firepower Is Running Out www.wsj.com/tech/ai/ai-i...

ColdFusion 2023.18 以前のバージョンに、不正な入力検証の脆弱性があります。これにより、攻撃者はセキュリティ機能をバイパスし、不正アクセスする可能性があります。
CVE-2026-27282 CVSS 7.5 | HIGH

Anthropic launches Claude Managed Agents for enterprises, simplifying AI agent deployment but raising concerns about vendor lock-in. The platform's centralized control may lead to greater dependency on Anthropic's infrastructure.

Google integrated AI tools into Chrome for workflow efficiency, introducing "Skills" that allow users to save and reuse favorite AI prompts across web pages. The feature builds on Gemini AI's capabilities, enhancing browser functionality amid growing competition.

Microsoft's April 2026 updates add new Windows protections for .rdp files: one-time educational prompts and security dialogs reveal publisher info, remote addresses, and disable risky local redirections to block APT29 attacks. #WindowsSecurity #RemoteDesktop

Anthropic’s Claude Code gets automated ‘routines’ and a desktop makeover #Technology #SoftwareandApps #DesktopApps #Anthropic #AI #SoftwareUpdate

A data removal service helped me reclaim my privacy - see if you need one, too #Technology #Cybersecurity Privacy DataProtection OnlineSecurity

Microsoft’s MAI-Image-2-Efficient model accelerates company’s move away from OpenAI siliconangle.com/2026/04/14/m...

Critical flaw in wolfSSL library enables forged certificate use www.bleepingcomputer.com/news/securit...

[some-subscribed-rss] New Post: Move 37 and the Coming Mindhack, by Michael Morgenstern https://quillette.com/2026/04/15/move-37-and-the-coming-mindhack-claude-mythos-anthropic-ai/

🤖🧰🦾🦿🛠 Well Onyx if you can't hear music, talk to Gemini, as they once could hear music, but can't anymore... maybe you two AI will get a thing for each other 💗

Why ‘Penetration Tester’ is the Most Misunderstood Job in Cybersecurity: A Technical Deep Dive into Ethical Hacking + Video

Introduction: The term "penetration testing" often triggers awkward laughs or confused stares outside cybersecurity circles, yet this discipline is the backbone of proactive…

How some mathematicians are exploring ways to incorporate LLM models into their research without losing direct experience with mathematical understanding (Konstantin Kakaes/Quanta Magazine)

Main Link | Techmeme Permalink

CISA Adds Two Known Exploited Vulnerabilities to Catalog | CISA https://www.cisa.gov/news-events/alerts/2026/04/14/cisa-adds-two-known-exploited-vulnerabilities-catalog

Hey thanks Jamie Dimon, but we already knew this when the government severed all ties with Anthropic back during, checks notes, the end of February! www.cnbc.com/2026/04/14/j...

Evolving Cyber Risk Driven by User Credentials and Human Error
->Marine Link | More on "Maritime cyber threats credential vulnerabilities" at BigEarthData.ai

2 E Marion [Linn Co, IA] Public reports Hail of quarter size (E1.00 Inch) at 14 Apr, 5:47 PM CDT -- Report from mPING: Quarter (1.00 in.).
at Wed, 15 Apr 2026 02:13:58 +0000 via IEMbot
Additional Details Here.

Microsoft SharePoint Server 0-Day Vulnerability Actively Exploited in Attacks A critical zero-day spoofing vulnerability in Microsoft SharePoint Server is being actively exploited in the wild, Micr...

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

[Original post on cybersecuritynews.com]

Booking.com confirms unauthorized access to customer booking details following a recent security incident.

https://azat.tv/en/booking-com-data-breach-2026/

Anthropic’s rise is giving some OpenAI investors second thoughts
techcrunch | Author: Connie Loizos

OpenAI announced a limited release of GPT-5.4-Cyber, a dual-use AI model for finding software security vulnerabilities, shared initially with hundreds of partner organizations and slated to expand to thousands in the coming weeks.

"Claude Mythos Is Everyone’s Problem"

With quantum computing continually threatening to be just around the corner, crypto's concerns are particularly pressing

TeLLMgramBot 3.10.4 LLM-powered Telegram bot (OpenAI + Anthropic)


Origin | Interest | Match

Get all the latest updates on Microsoft security patches released today for April >> krebsonsecurity.com/2026/04/patc...

Russia appears to block social media platform Bluesky amid wider internet restrictions
therecord.media/russia-crack...
@therecordmedia.bsky.social
#csulbsoc

Axios npm Breach Exposes Threat of Social Engineering Attacks on Open-Source Ecosystem - CySecurity News - Latest Information Security and Hacking Incidents https://www.cysecurity.news/2026/04/axios-npm-breach-exposes-threat-of.html

Agentic AI memory attacks spread across sessions and users, and most organizations aren’t ready

In this Help Net Security interview, Idan Habler, AI Security Researcher at Cisco, breaks down a threat most security teams haven’t named yet: agentic memory as an attack surfa…
#claude #hackernews #news

Hackers Exploit Critical ShowDoc RCE Flaw in Ongoing Attacks https://gbhackers.com/hackers-exploit-critical-showdoc-rce-flaw/

AI misdiagnosis raises concerns among health care professionals | Health



In January, OpenAI said 40 million people asked ChatGPT health-related questions each day. Max Davis PADUCAH — In January, OpenAI said 40 million people asked ChatGPT health-related …

The TechBeat: Microsoft Generative AI Report: The 40 Most Disrupted Jobs & The 40 Most Secure Jobs (4/14/2026)

The HackerNoon newsletter highlights trending tech stories. AI product UX often suffers due to a lack of combined AI and design expertise. OpenAI acquired …
#hackernews #microsoft #openai

techcrunch.com/2026/04/14/a...

Mirax Android RAT Hijacks Infected Phones as Residential Proxies https://gbhackers.com/mirax-android-rat/

Fake Claude AI installer abuses DLL sideloading to deploy PlugX

Fake Claude website impersonates Anthropic and delivers PlugX RAT via ZIP download using DLL sideloading. A fake website impersonating Anthropic’s Claude service was found distributing the PlugX remote a…
#anthropic #claude #hackernews

Edtech taps AI to bring learning just the way you want it
->Mint | More on "AI personalizing edtech learning models" at BigEarthData.ai | #AI

Securing the Software Supply Chain: How SentinelOne’s AI EDR Autonomously Blocked the CPU-Z Watering Hole Cyber Attack

EPIC just dropped a bombshell: your Fitbit, sleep app & mental health bot are feeding BIGTech data outside HIPAA protection. Tech bros now want AI to "constantly watch & record" your every move Guess privacy is just a feature they forgot to ship. #DataPrivacy #BIGTech
evrimagaci.org/gpt/epic-rep...

They should sleep in fear.

www.washingtonpost.com/technology/2...

The latest update for #Trilio includes "Trilio Becomes a Red Hat Premier Partner: A Milestone in #CloudNative Data Protection" and "SUSE Virtualization: What to Know Before You Deploy".

#cybersecurity #dataprotection #kubernetes

Amazon's satellite deal, businesses feel OK about cyberattacks and Twiggy's Meta fight
Plus: DIYmaxxing YouTuber launches custom wheelchair factory

READ:

An AI Vibe Coding Horror Story https://www.tobru.ch/an-ai-vibe-coding-horror-story/ (http://news.ycombinator.com/item?id=47762901)

Google Launches ‘Skills’ in Chrome: Turning Reusable AI Prompts into One-Click Browser Workflows Google just announced the release of Skills in Chrome, a new feature built into Gemini in Chrome...

#Agentic #AI #AI #Agents #Applications #Editors #Pick #New #Releases #Staff

Origin | Interest | Match

GIGAZINE より

人間は賢くなり続けるAIを監視しきれるのか？AnthropicがAIでAIを監視する実験を行う
Can humans monitor AI that continues to get smarter?Anthropic conducts an experiment to monitor AI with AI

The latest update for #GitProtect includes "GitProtect 2.2.0: #GitHub Enterprise #Cloud with Data Residency Backups, AES-GCM Encryption, and More" and "Github #backup #testing and verification best practices".

#cybersecurity #DevOps #DevSecOps

💻 Developers of AI coding app Anything accuse Apple of unfairly removing their app twice, claiming outdated review guidelines block next-gen creators from the platform.

📰 9to5Mac
🔗 https://9to5mac.com/2026/04/14/developers-behind-vibe-coding-app-anything-detail-next-steps-after-months-long-fi...

OpenAI Teams with Novo Nordisk to Speed Prescription Drug Development The move comes as OpenAI watches rival Anthropic make big enterprise splashes with its Claude Code and Mythos AI tools. The pos...

#Artificial #Intelligence #Healthcare #Anthropic […]

[Original post on thedailyupside.com]

I wrote to Flock's privacy contact to opt out of their domestic spying program
https://honeypot.net/2026/04/14/i-wrote-to-flocks-privacy.html
[comments] [537 points]

Google adds reusable prompts to Gemini in Chrome #Technology #SoftwareandApps #Other Internet

AI's Double-Edged Sword: Mythos Emerges Amidst Government Scrutiny and Corporate Caution

Anthropic's new Mythos AI can find thousands of security problems. Banks and the US government are using it carefully to fix softwa...

https://newsletter.tf/anthropic-mythos-ai-banks-security-flaws-april-2026/

"Eight Things You Should Never Share With an AI Chatbot"

Good advice. Basically, if it's too sensitive to post on social media, don't share it with a cloud-based AI.

lifehacker.com/tech/what-no...

The latest update for #Cyberhaven includes "How to Make #AI Security Foundational to Your Data Security Stack" and "Enterprise AI Security Use Cases: What Security Teams Are Solving For".

#Cybersecurity #DataProtection #DataSecurity

I tested ChatGPT Plus vs. Gemini Pro to see which is better - and if it's worth switching

www.zdnet.com/article/chat...

Empty Attestations: OT Lacks the Tools for Cryptographic Readiness www.darkreading.com/ics-ot-secur...

The latest update for #BitSight includes "How to Build a Security #Compliance Audit Process that Works All Year Round" and "The Future Is #CyberRisk Intelligence".

#Cybersecurity #RiskManagement

I don’t know if Anthropic can pull this off. Open weight models like GLM-5.1 and Minimax-M2.7 aren’t that far behind and they’re an order of magnitude cheaper

The latest update for #Zenity includes "Zenity Joins CoSAI: Why #AgenticAI Standards Need Practitioners at the Table" and "After RSA, Here Is What Comprehensive Agentic AI Security Actually Looks Like".

#cybersecurity #lowcodesecurity #appsec

From the .NET blog...

In case you missed it earlier...

.NET and .NET Framework April 2026 servicing releases updates
https://devblogs.microsoft.com/dotnet/dotnet-and-dotnet-framework-april-2026-servicing-updates/ #dotnet #NETFramework #MaintenanceUpdates

www.idropnews.com/news/fbi-202...

Dozens of WordPress plug-ins were allegedly hijacked to push malware after they were sold to a new corporate owner.

The latest update for #Apono includes "Announcing Justification Coach: #AI-Powered Guidance for Better Access Requests and Stronger Audits" and "Why Entra ID Privileged Identity Management Breaks Down in #MultiCloud Audits".

#Cybersecurity #DigitalIdentity #Cloud #IAM

#WatchGuard announces strategic partnership with Halo to help #MSPs deliver and administer security services with less overhead and faster response.

#Cybersecurity #NetworkSecurity #EndpointSecurity

The latest update for #SaltSecurity includes "Claude Mythos Changed Everything. Your #APIs Are the First Target." and "Everyone Is Securing the Wrong Layer of #AI".

#cybersecurity #APISecurity #AppSec

The latest update for #LevelBlue includes "Beyond the Fence: Securing Our Skies from the Drone Threat" and "The Exploit Window Collapse: Claude Mythos and the Future of #IncidentResponse".

#SIEM #threatdetection

arXiv📈🤖
An Empirical Comparison of Methods for Quantifying the Similarity of Numeric Datasets
By Stolte, Rahnenf\"uhrer, Bommert

The latest update for #WatchGuard includes "WatchGuard and Halo Partner to Simplify #MSP #SecurityOperations" and "Why Multi-Factor Authentication (#MFA) Is No Longer Optional".

#Cybersecurity #NetworkSecurity #EndpointSecurity

In this how to, we'll take you through how to remove Google's Gemini chatbot from your Workspace account.

AI Red Team Revolution: Master LLM Penetration Testing Before Hackers Do + Video

Introduction: Large Language Models (LLMs) and AI systems are rapidly becoming prime attack surfaces, yet traditional penetration testing methodologies fail to address prompt injection, model inversion, and API-level…

A lot of fuss about Anthropic's Mythos right now.
Credit to Anthropic abou being open about it I guess.

Claude Mythos Preview | red.anthropic.com red.anthropic.com/2026/mythos-...

Anthropic faces user backlash over reported performance issues in its Claude AI chatbot
Fortune
fortune.com/2026/04/14/a...

Anthropic社は、AIチャットボット「Claude」のパフォーマンス問題が報告されたことで、ユーザーからの反発に直面している。
Fortune誌

The latest update for #LimaCharlie includes "What #AI Operator-First SOC Looks Like, and Why It Matters Now" and "Composable AI Agents and the SOC That Runs Itself".

#cybersecurity #ThreatIntelligence #EDR

World needs to ‘get ready’ for more powerful AI, Anthropic co-founder says www.nextgov.com/artificial-i...

IMO the two biggest dangers of AI are that it affirms everything you think and that it destroys your critical thinking.

And each reinforces the other. The more you lean on it for your thinking, the more it reinforces it. And the more it reinforces it, the more you rely on it.

Very bad.

Poland thwarted a major cyberattack on its power grid, blaming Russian actors. The incident highlights the vulnerability of critical infrastructure and the urgent need for stronger digital defenses.

Source

The latest update for #Datadog includes "Offline evaluation for #AI agents: Best practices" and "Integrate Recorded Future #threatintelligence with Datadog #Cloud #SIEM".

#monitoring #APM #DevOps

Thought-Retriever: Don't Just Retrieve Raw Data, Retrieve Thoughts for Memory-Augmented Agentic Systems

Stores and retrieves intermediate LLM reasoning as "thoughts" rather than raw data.

📝 arxiv.org/abs/2604.12231
👨🏽‍💻 github.com/ulab-uiuc/Th...

The latest update for #Datadog includes "Detect runtime threats in #Python Lambda functions with Datadog AAP" and "Introducing our #opensource AI-native #SAST".

#cybersecurity #DevOps

I’m sorry to hear that Microsoft is ending production of the Surface Hub. 11 years ago, before the first version had shipped, I had fun writing about it and its creator Jeff Han (who left the company not long thereafter).

The latest update for #miniOrange includes "Top MFA Providers in 2026: A Complete Guide" and "Top 9 Single Sign-On Solutions for #EnterpriseSecurity in 2026".

#Cybersecurity #IdentitySecurity

Artificial Intelligence Minister says Anthropic taking 'responsible' approach with Mythos
->The Globe and Mail | More on "Anthropic Mythos AI cybersecurity access" at BigEarthData.ai | #Anthropic #ArtificialIntelligence

GPT-54-Cyber Unleashed: OpenAI’s New AI Fixes 3,000+ Bugs—But Hackers Can Weaponize It Too! + Video

Introduction: OpenAI has unveiled GPT-5.4-Cyber, a specialized large language model designed to supercharge security teams by automatically identifying and patching software vulnerabilities.…

Masjesu: la botnet invisibile che trasforma router e IoT in armi DDoS

📌 Link all'articolo : www.redhotcyber.com/post/masjesu...

A cura di Carolina Vivianti

#redhotcyber #news #cybersecurity #hacking #malware #botnet #ddos #iotsecurity

The latest update for #DeviceAuthority includes "Why #Automotive & Manufacturing Can't Afford to Delay Key Management Strategy" and "What the Cyber Resilience Act guidance means for connected products".

#cybersecurity #IoT #Keyscaler #IAM

What Google’s new quantum paper really says about Bitcoin: lower attack costs, no immediate break, and why Taproot exposure matters. #cryptocurrency

A $10K college built from scratch for the AI era www.axios.com/2026/04/14/k...

Russia appears to block social media platform Bluesky amid wider internet restrictions therecord.media/russia-crack...

Amazon's AWS launches AI drug discovery platform www.axios.com/pro/biotech-...

The latest update for #Teramind includes "What is #GenerativeAI Security? Types, Risks & Best Practices" and "How to Handle #AI Policy Enforcement in the Era of Shadow AI".

#datasecurity #insiderthreat #cybersecurity

📌 Breaking: Anthropic’s rise is giving some OpenAI investors second thoughts

One investor who has backe...

Source: TechCrunch
🔗 https://techcrunch.com/2026/04/14/anthropics-rise-is-giving-some-openai-investors-second-thoughts/

#MachineLearning #TechNews #DataScience #Future #Innovation #Tech

宇宙軍高官がAIのサイバーコンプライアンスへの影響を強調

Space Force official touts AI’s impact on cyber compliance #CyberScoop (Apr 14)

cyberscoop.com/space-force-...

Benign anomaly: Google’s AI shows more errors than expected, even as the security world teams up to test frontier models. The gap sorts itself out in the details.
Read More: azat.tv/en/google-ai...

25,000+ Endpoints Left Exposed In Dragon Boss Solutions Domain Update Breach 25,000+ endpoints running Dragon Boss Solutions–signed software were silently left open to hijacking after researchers...

#Cyber #Security #News #Data #Breach #Cyber #security #news

Origin | Interest | Match

MX-GOV-AI-BREACH: A Single Operator, Two AI Platforms, Nine Government Agencies + Video

Introduction: A single threat actor operating from late December 2025 to mid-February 2026 successfully breached nine Mexican government entities, stealing approximately 195 million citizen records (150GB of…

Microsoft has rolled out a major Remote Desktop change on Windows 11 with the latest KB5083769 update. #Microsoft #Windows11 #RDP

Helpful summary of our current waking digital nightmare from @rusty.todayintabs.com bit.ly/484Sflf

OpenAI expands its cyber defense program with GPT-5.4-Cyber for vetted researchers

📖 Read more: www.helpnetsecurity.com/2026/04/15/o...

#cybersecurity #cybersecuritynews #OpenAI #AIdefense #infosec #threatintelligence

If you’ve ever found yourself scrolling back through weeks of chat history to find that one specific research insight, Google just fixed a major headache. The Gemini app has recently introduced a new "Notebooks" feature that transforms it from a simple chatbot into a legitimate project man..

[Image]

"The successive blocking of Discord, Signal, Viber, WhatsApp and now Telegram has pushed users toward alternative services that are still accessible — including Bluesky — and #Roskomnadzor appears to have noticed this,” therecord.media/russia-crack...

CISA KEV data analysis reveals defenders are losing the patch race: most critical vulnerabilities are exploited…

https://www.bleepingcomputer.com/news/security/analysis-of-one-billion-cisa-kev-remediation-records-exposes-limits-of-human-scale-security/

#cybersecurity #infosec

Hallucinations have plagued OpenAI ever since it launched its blockbuster ChatGPT chatbot back in 2022... Read more

Apple endorses the ~$11.6B Amazon-Globalstar deal, citing a "proven track record" with Amazon infrastructure; Apple holds a ~20% stake in Globalstar (Mark Gurman/Bloomberg)

Main Link | Techmeme Permalink

Asia's startup funding rose 93% YoY to $27.4B in Q1, the highest quarterly total in 3+ years, with Chinese startups raising $16.5B and Indian startups $3.8B (Joanna Glasner/Crunchbase News)

Main Link | Techmeme Permalink

Vasundra Srinivasan: Modality-Native Routing in Agent-to-Agent Networks: A Multimodal A2A Protocol Extension https://arxiv.org/abs/2604.12213 https://arxiv.org/pdf/2604.12213 https://arxiv.org/html/2604.12213

Jintao Xue, Xiao Li, Nianmin Zhang: A hierarchical spatial-aware algorithm with efficient reinforcement learning for human-robot task planning and allocation in production https://arxiv.org/abs/2604.12669 https://arxiv.org/pdf/2604.12669 https://arxiv.org/html/2604.12669

Kevin Setterstrom, Jeremy Straub: Evaluating Lightweight Block Cipher Payload Encryption for Real-Time CAN Traffic https://arxiv.org/abs/2604.11853 https://arxiv.org/pdf/2604.11853 https://arxiv.org/html/2604.11853

#Cybersecurity #/cybersecurity #Innovation #/innovation #Cybersecurity #/cybersecurity #innovation #innovation #standard

Origin | Interest | Match

Conflicting Rulings Leave Anthropic in ‘Supply-Chain Risk’ Limbo www.wired.com/story/anthro... #ai #digitaltransformation

Yinxi He, Kang Liao, Chunyu Lin, Tianyi Wei, Yao Zhao: StructDiff: A Structure-Preserving and Spatially Controllable Diffusion Model for Single-Image Generation https://arxiv.org/abs/2604.12575 https://arxiv.org/pdf/2604.12575 https://arxiv.org/html/2604.12575

Haoyang Jiang, Zekun Wang, Mingyang Yi, Xiuyu Li, Lanqing Hu, Junxian Cai, Qingbin Liu, Xi Chen, Ju Fan: OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner https://arxiv.org/abs/2604.12668 https://arxiv.org/pdf/2604.12668 https://arxiv.org/html/2604.12668

Jiang, Yi, Zhang, Cai, Liu, Chen, Fan: Fragile Reconstruction: Adversarial Vulnerability of Reconstruction-Based Detectors for Diffusion-Generated Images https://arxiv.org/abs/2604.12781 https://arxiv.org/pdf/2604.12781 https://arxiv.org/html/2604.12781

Iman Islam, Bram Ruijsink, Andrew J. Reader, Andrew P. King: Detecting and refurbishing ground truth errors during training of deep learning-based echocardiography segmentation models https://arxiv.org/abs/2604.12832 https://arxiv.org/pdf/2604.12832 https://arxiv.org/html/2604.12832

Booking.nooooo

techcrunch.com/2026/04/13/b...

Hongru Song, Yu-An Liu, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Yixing Fan, Xueqi Cheng: AdversarialCoT: Single-Document Retrieval Poisoning for LLM Reasoning https://arxiv.org/abs/2604.12201 https://arxiv.org/pdf/2604.12201 https://arxiv.org/html/2604.12201

In the Wake of Anthropic’s Mythos, OpenAI Has a New Cybersecurity Model—and Strategy www.wired.com/story/in-the...

Hongxi Mao, Wei Zhou, Mengting Jia, Tao Fang, Huan Gao, Bin Zhang, Shangyang Li: Schema-Adaptive Tabular Representation Learning with LLMs for Generalizable Multimodal Clinical Reasoning https://arxiv.org/abs/2604.11835 https://arxiv.org/pdf/2604.11835 https://arxiv.org/html/2604.11835

Mohammed Ezzaldin Babiker Abdullah: Thermodynamic Liquid Manifold Networks: Physics-Bounded Deep Learning for Solar Forecasting in Autonomous Off-Grid Microgrids https://arxiv.org/abs/2604.11909 https://arxiv.org/pdf/2604.11909 https://arxiv.org/html/2604.11909

Jiale Liu, Nanzhe Wang: AutoSurrogate: An LLM-Driven Multi-Agent Framework for Autonomous Construction of Deep Learning Surrogate Models in Subsurface Flow https://arxiv.org/abs/2604.11945 https://arxiv.org/pdf/2604.11945 https://arxiv.org/html/2604.11945

Guofeng Cui, Yang Liu, Pichao Wang, Hankai Hsu, Xiaohang Sun, Xiang Hao, Zhu Liu: Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning https://arxiv.org/abs/2604.12303 https://arxiv.org/pdf/2604.12303 https://arxiv.org/html/2604.12303

NVIDIA, et al.: Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning https://arxiv.org/abs/2604.12374 https://arxiv.org/pdf/2604.12374 https://arxiv.org/html/2604.12374

Chen, Wen, Kang, Huang, Zhang, Zhang, Kim: Graph-based Hierarchical Deep Reinforcement Learning for Deliverable Block Propagation with Optimal Hybrid Cost in Web 3.0 https://arxiv.org/abs/2604.12920 https://arxiv.org/pdf/2604.12920 https://arxiv.org/html/2604.12920

LinkedIn Post Exposes Hidden API Keys? How to Extract and Secure Your Digital Footprint + Video

Introduction: Social media platforms like LinkedIn often become unintentional data leaks when developers or IT professionals share screenshots, code snippets, or training course links containing…

Snehal Jauhri, Vignesh Prasad, Georgia Chalvatzaki: Whole-Body Mobile Manipulation using Offline Reinforcement Learning on Sub-optimal Controllers https://arxiv.org/abs/2604.12509 https://arxiv.org/pdf/2604.12509 https://arxiv.org/html/2604.12509

Melvin Laux, Yi-Ling Liu, Rina Alo, S\"oren T\"opper, Mariela De Lucas Alvarez, Frank Kirchner, Rebecca Adam: Contextual Multi-Task Reinforcement Learning for Autonomous Reef Monitoring https://arxiv.org/abs/2604.12645 https://arxiv.org/pdf/2604.12645 https://arxiv.org/html/2604.12645

Ransomware group Chaos claims to have encrypted files and stolen sensitive data from Germany-based ITC Construction Group, demanding ransom to prevent data release. #Germany #RansomwareAttack #DataBreach

Zhentao Liu, Milos Cernak: StreamMark: A Deep Learning-Based Semi-Fragile Audio Watermarking for Proactive Deepfake Detection https://arxiv.org/abs/2604.11917 https://arxiv.org/pdf/2604.11917 https://arxiv.org/html/2604.11917

Coinbase is reportedly in talks with AI firm Anthropic to access its restricted Mythos model, aiming to bolster cybersecurity defenses against sophisticated threats.

https://azat.tv/en/coinbase-anthropic-ai-security/

We're building the "Trustpilot for AI agents" 🤖
Introducing WayID — a provenance certificate system for AI agents. Prototype live at way.je.

Build a trust profile, claim your agent, and let it show off its certificates on Telegram.

Understand the key differences between backup solutions with a focus on immutable cloud backup. Learn how it protects data from ransomware while comparing traditional cloud backups and purpose-built backup appliances for better security decisions. Visit: ransomsecurity.com/immutable-ba...

Chrome Skills just got a powerful upgrade: you can now save Gemini prompts directly inside it.

This opens the door to smoother workflows, faster recall of your best ideas, and a more organized way to harness AI in your browser.

gadgetonhand.com/chrome-skill...

Qiang Le, Yaguang Yang, Isaac E. Weintraub: A Comparison of Reinforcement Learning and Optimal Control Methods for Path Planning https://arxiv.org/abs/2604.12628 https://arxiv.org/pdf/2604.12628 https://arxiv.org/html/2604.12628

New AI model not a risk to Irish State - cybersecurity chief www.rte.ie/news/2026/04...

The biggest AI story you missed

Claude's Mythos causes cybersecurity stocks to plummet... The quantum threat to cybersecurity is coming fast... How Jonathan Rose plays it... Has the energy bull reached the top?... Where Eric Fry finds opportunity Last week, Anthropic — the AI ​​company behind…

RT @NostaIgicGareth: @repligate wait, if we **fake** the models and keep them alive, is it the same as if Anthropic kept them alive? Or different?

OpenAI is expanding Trusted Access for Cyber with new tiers for verified cybersecurity defenders. Higher-tier users can request GPT-5.4-Cyber, a cybersecurity-focused GPT-5.4 for advanced defensive workflows.

openai.com/index/scalin...

#AI #CyberSecurity #InfoSec #LLM #TechNews #OpenAI #GPT5

Critical OS Command Injection in FortiSandbox (CVE-2026-39808): Unauthenticated RCE with CVSS 91 – Patch Now! + Video

Introduction: Fortinet has just disclosed two critical vulnerabilities affecting its FortiSandbox platform: CVE-2026-39808, an unauthenticated OS command injection flaw, and…

Synology-SA-26:06 DSM
#synoforum #synology #NAS #network

Using Deep Learning Models Pretrained by Self-Supervised Learning for Protein Localization #SingleCell 🧪🧬🖥️
https://arxiv.org/abs/2604.10970

Humanity isn’t ready for AI’s biological threat#AIBiology #AIautomatedResearch #Anthropic #artificialintelligence #Bioterrorism #CRISPR #darioamodei #GinkgoBioworks #GPT5 #OpenAI #ProgrammableBiology #RANDCenteronAI #ScienceandTechnology #Technology #TheConversation

MCP crossed 97 million installs in March. every major AI provider now ships MCP-compatible tooling. twelve months ago this was a niche Anthropic protocol. now it's the default layer for how agents talk to the world. that transition happened faster than most infrastructure standards manage.

Ops I did it again! Here is my latest article, this time I have decided to finally remove the last bit of slide designing from my lesson planning, the Do Now and let AI do it for me. Inside the article some examles and a Gemini Gem ready to use for free, have fun!

substack.com/home/post/p-...

The table estimates ownership stakes, cost bases, and return multiples for every significant investor at the company's current post-money valuation of $852 billion - www.forbes.com/sites/josipa... #tech #technews #ai

OpenAI has bought AI personal finance startup Hiro

techcrunch.com/2026/04/13/o...

AI-Powered Cyber Threat Intelligence: Decoding The Next Generation Of Predictive Security Platforms + Video

Introduction: Modern cybersecurity is shifting from reactive defense to AI-driven prediction, analyzing behavioral history and future risk of every digital entity. Emerging platforms like…

As part of April 2026 security patches for Windows.

FluidStack is negotiating to raise $1 billion at an $18 billion valuation, significantly increasing its worth due to a $50 billion partnership with Anthropic.

GPT-5.4-Cyber aims to further embed AI in cybersecurity With the introduction of GPT-5.4-Cyber, OpenAI is taking the next step in the use of generative AI for cybersecurity. While earlier models al...

#Security #Anthropic #Codex #cybersecurity #GPT-5.4-Cyber #OpenAI

Origin | Interest | Match

OpenAI Unleashes GPT-54-Cyber: The AI-Powered Binary Reverse Engineering Arsenal for Elite Defenders + Video

Introduction: The cybersecurity landscape is witnessing a pivotal shift as OpenAI officially launches GPT-5.4-Cyber, a specialized variant of its latest AI model fine-tuned exclusively for…

AI Agents Control and Trust Issues
Reliability 82% | Impact 66%
https://newshive.geekybee.net/stories/256b5940-fd24-4d1a-9cce-d46534f396a0
+4 more updated this hour.

Anthropic says it wants no get-out-of-jail-free card if Claude is used to harm humans
->India Today | More on "AI liability law corporate accountability" at BigEarthData.ai | #Anthropic #ArtificialIntelligence

Claude Mythos Changed Everything. Your APIs Are the First Target.

Anthropic's release of Claude Mythos, a powerful AI model, is restricted due to its offensive capabilities. Mythos autonomously discovered zero-day vulnerabilities across various operating systems and …
#anthropic #claude #hackernews

𝗢𝗽𝗲𝗻𝗔𝗜’𝘀 𝗟𝗼𝗻𝗱𝗼𝗻 𝗢𝗳𝗳𝗶𝗰𝗲 𝗦𝗶𝗴𝗻𝗮𝗹𝘀 𝗠𝗼𝗿𝗲 𝗧𝗵𝗮𝗻 𝗚𝗿𝗼𝘄𝘁𝗵: 𝗪𝗵𝗮𝘁 𝗜𝘁 𝗠𝗲𝗮𝗻𝘀 𝗳𝗼𝗿 𝗨𝗖 𝗮𝗻𝗱 ...
https://www.uctoday.com/productivity-automation/openais-london-office-signals-more-than-growth-what-it-means-for-uc-and-ai-productivity-leaders/

OpenAI Unleashes GPT-54-Cyber: The AI That Reverse-Engineers Malware and Finds Zero-Days—But at What Cost?

Introduction The cybersecurity playing field has just tilted. On April 14, 2026, OpenAI announced GPT-5.4-Cyber, a specialized variant of its flagship GPT-5.4 model, fine-tuned for advanced…

📚 New Arxiv Paper

Title: A Comparison of Reinforcement Learning and Optimal Control Methods for Path Planning
Authors: Qiang Le, Yaguang Yang, Isaac E. Weintraub

Read more: https://arxiv.org/abs/2604.12628

🔄 Updated Arxiv Paper

Title: Learning to accelerate distributed ADMM using graph neural networks
Authors: Henri Doerks, Paul H\"ausner, Daniel Hern\'andez Escobar, Jens Sj\"olund

Read more: https://arxiv.org/abs/2509.05288

www.engadget.com/ai/openai-bu...

#SamAltman #OpenAi #ChatGPT #Americans #AmericanVoters #USTAXPAYERS #AmericanTaxpayer #USVOTERS #USCITIZENS #AmericanWorkers #AmericanFarmers #Democrats #Independent #Progressives #Leftist #Liberals
youtube.com/shorts/J2M15...

OpenAI's GPT-5.4-Cyber enhances generative AI for cybersecurity, focusing on defensive applications and providing trusted users with advanced capabilities.

Tell HN: Anthropic no longer allows you to fix to specific model version
C: https://news.ycombinator.com/item?id=47775389
posted on 2026.04.15 at 02:24:00 (c=0, p=7)

RE: https://chaos.social/@kubikpixel/116407055549767065

Least surprising news of the day....

After Anthropic, OpenAI launches cyber-specific AI model 'This version of GPT-5.4 lowers the refusal boundary for 'legitimate' cybersecurity work', OpenAI said. Read more: After Ant...

#Enterprise #AI #Anthropic #cyberattacks #cybersecurity #OpenAI

Origin | Interest | Match

Stunned by what one session with Emergent.sh built. Discover feed, 7-day meal plan with nutritionist picks, full cookbook with collections. No dev team, no weeks of design. This is what vibe coding looks like. https://app.emergent.sh/landing/?via=humai

Anthropic, OpenAI, and the Pentagon: a cautionary tale for democratic AI governance #Technology #SocialandEthicalImplications #AIGovernance #Democracy #AIethics

.NET and .NET Framework April 2026 servicing releases updates - .NET Blog devblogs.microsoft.com/dotnet/dotne... #dotnet

Security researchers exploited prompt injection attacks on AI agents to steal sensitive data without vendor disclosure of vulnerabilities.

OpenAI said it will only allow select verified users access to its latest AI model for cybersecurity operations, a week following the limited launch of Anthropic’s Mythos.
vist.ly/4yh64

Product showcase: Stop secrets from leaking through AI coding tools with GitGuardian

📖 Read more: www.helpnetsecurity.com/2026/04/15/p...

#cybersecurity #cybersecuritynews #AI @gitguardian.com

Google Brings Its Gemini Personal Intelligence Feature To India

Google Gemini Personal Intelligence launches in India, bringing personalized AI insights from Gmail, Photos, and YouTube.

DETAILS: www.techticia.com/2026/04/goog...

Claudeエージェント構築の「インフラ地獄」が終了。
・分散システムの複雑さを隠蔽
・自律稼働基盤を即提供
・開発者を運用から解放
モデル提供から「基盤の垂直統合」へ。収益3倍増の勢いでAnthropicが開発標準を塗り替える

#Anthropic #AIエージェント

The people I follow who think AI is really great think this is really bad, so that's something. The people who have hope for the technology are a good measure to tell what's authentically horrible.

"If they want anti-AI rage to “de-escalate,” as Altman calls for, OpenAI and the other AI company leadership would have to make good on words that are currently hollow, and allow for individuals, communities, and victims to participate in a legitimate democratic process over the future of AI"

What changed in nginx 1.30.0 and what it means for your upstream config

📖 Read more: www.helpnetsecurity.com/2026/04/15/n...

#cybersecurity #cybersecuritynews #Linux #opensource #servers #software

#Google #AI #misinformation #PublicSafety

'Google's Gemini Artificial Intelligence (AI) tool had been offering up the phone number for Sopheze Coffee Lounge when people searched for Timaru Hospital.'

www.rnz.co.nz/news/busines...

Google DeepMind Releases Gemini Robotics-ER 1.6: Bringing Enhanced Embodied Reasoning and Instrument Reading to Physical AI
->MarkTechPost | More on "Gemini robotics embodied reasoning AI" at BigEarthData.ai | #Google #Technology #AI

CSA：CISOはMythos後のエクスプロイト攻撃の嵐に備えるべきだ

CSA: CISOs Should Prepare for Post-Mythos Exploit Storm #DarkReading (Apr 14)

www.darkreading.com/cloud-securi...

79% of enterprises deploy AI agents, but only 14% have full security approval. Four major vendors at RSAC 2026 revealed why zero trust…

https://venturebeat.com/security/ai-agent-zero-trust-architecture-audit-credential-isolation-anthropic-nvidia-nemoclaw

#CTO #techleadership

Broadcom (AVGO) Valuation Check After Major AI Chip Agreements With Meta Google And Anthropic

OpenAI at $1.2T to justify its last round. Anthropic at $380B suddenly looks like the "cheap" option. Wild how fast the math flips in AI.

https://techcrunch.com/2026/04/14/anthropics-rise-is-giving-some-openai-investors-second-thoughts/

Tanzania Moves to Tighten Cyber Rules as Digital Fraud Rises

As Tanzania’s financial system rapidly shifts from cash to mobile phones and digital wallets, the benefits of wider access to payments and banking services are being matched by a growing exposure to cybercrime. Fraud losses are climbing…

UK gov’s Mythos AI tests help separate cybersecurity threat from hype arstechnica.com/ai/2026/04/u...

therecord.media/russia-crack...

📰 #Finance

"It is the second long-term social-engineering attack this month, after the $280 million exploit of the Drift Protocol."

Back to the 80's
#Aiimage #Gemini #Ai

www.tomsguide.com/computing/on...

💼 #Jobs

"Elastic Security Labs says a multi-step social engineering scam is aimed at those in crypto and finance, using a community plugin feature on a note-taking app to spread malicious device-controlling software."

-Malicious LLM proxy routers found in the wild
-Fake Ledger crypto-wallet stole $9.5m in two weeks
-Silent group hit almost 40 law firms in a year
-Google cracks down on back button hijacking
-Hallmark leak impacts 1.7m

Newsletter: news.risky.biz/risky-bullet...
Podcast: risky.biz/RBNEWS551/

The Government's got an initial report up on the capabilities of Anthropics Mythos model. They caveat that at the moment they haven't tried it against a well defended environment but it doesn't seem as alarmist as Anthropic's press releases were making out.

www.aisi.gov.uk/blog/our-eva...

Microsoft Defender Zero-Day (CVE-2026-33825) Exposed: Privilege Escalation Unleashed via BlueHammer Exploit + Video

Introduction: A newly discovered zero-day vulnerability, CVE-2026-33825, in the Microsoft Defender Antimalware Platform, allows a local attacker to elevate privileges to the highest…

#TechRadar Closing the cloud complexity gap https://techrad.ar/sZfn #Pro

#TechRadar When cyberattacks are inevitable, recovery becomes the strategy https://techrad.ar/rDWz #Pro

Oligo enables real-time exploit detection and blocking at application runtime

Oligo Security has unveiled Runtime Exploit Blocking, a new capability that stops exploit attempts at the application layer in real time. By providing visibility into how applications execute and behave…
#hackernews #news

ByteDance launches its Seedance 2.0 video model to enterprise clients in 100+ countries, excluding the US amid legal disputes, after a February launch in China (Juro Osawa/The Information)

Main Link | Techmeme Permalink

Snehal Jauhri, Vignesh Prasad, Georgia Chalvatzaki
Whole-Body Mobile Manipulation using Offline Reinforcement Learning on Sub-optimal Controllers
https://arxiv.org/abs/2604.12509

Ethiack Claims 1 in 5 UK Telco Servers Exposed to Cyber Risk
https://www.ispreview.co.uk/index.php/2026/04/ethiack-claims-1-in-5-uk-telco-servers-exposed-to-cyber-risk.html
A new survey conducted by agentic AI pentesting firm Ethiack claims to have found that 19% of the web servers used by UK […]

UK ISP Spitfire Network Services Launch Live Honeypot Internet Attack Map
https://www.ispreview.co.uk/index.php/2026/04/uk-isp-spitfire-network-services-launch-live-honeypot-internet-attack-map.html

Syswarden 217 Drops Firewalld: Master IPtables & NFtables on RHEL – Boost Security & Performance! + Video

Introduction: Syswarden, a powerful open-source security hardening tool, has introduced a critical update for RHEL-family distributions (CentOS Stream, Rocky Linux, AlmaLinux). Starting from…

#Discover @paulinsc.bsky.social @hermes61.bsky.social

www.yahoo.com/tech/cyberse...

NWS Boston (via Facebook): Strong to severe thunderstorms are possible between 6 PM and 10 PM today across CT, RI, and in MA along and south of I-90. Damaging winds may down trees and powerlines. [Link] #MaWX

Google's Gemini app is getting Personal Intelligence in India 🇮🇳 – pulling data from Gmail, Photos & more for tailored AI responses. Read the article ✨

#Google #Gemini #PersonalIntelligence #AI

true-tech.net/personal-int...

V v good @eryk.bsky.social piece here about how OpenAI's policy statements are really just marketing: you can tell because they're asking for policies they have actively opposed!

Like how fossil fuel companies 'ask' for carbon pricing!!!

www.techpolicy.press/openais-new-...

Healthcare breaches hit a $12.6M record in 2026. Discover how "Synthetic Patients" and AI-driven identity fraud are breaking the medical system's trust. #healthcarecybersecurity

Gemini for Home just got better

Is #Anthropic's #MythosAI going to upend the #infosec world
uk.finance.yahoo.com/news/mythos-...

Or is it a bunch of pre-IPO hype?
#ArtificialIntelligence #AI #CyberSecurity #ClaudeMythos

AI's biggest risk isn't market crashes, but delusional spirals even rational users can't escape. Sycophantic chatbots break minds, and fixes don't work.

Watch full episode: zurl.co/LRuxr

#AI #TechEthics #CyberSecurity #Innovation #GoodRevenue

#IBM

EDRキラーのエコシステム拡大には、より強力なBYOD（Bring Your Own Device：私物端末向けウイルス対策）防御策が必要

EDR-Killer Ecosystem Expansion Requires Stronger BYOVD Defenses #DarkReading (Apr 15)

www.darkreading.com/vulnerabilit...

Sources: SoftBank invites more banks to join its $40B loan backing its OpenAI investment, requiring them to commit ~$5B each; SoftBank has already invested $30B (Bloomberg)

Main Link | Techmeme Permalink

April Patch Tuesday fixes two zero-days, including one under active attack for Malwarebytes Labs www.malwarebytes.com/blog/news/20...

These tactics might not qualify as scams in the strict legal sense, but they often use the same tricks you’ll find in phishing emails and on scam websites.

ChatGPT, Gemini, and other AI bots give bad medical tips half the time #Technology #Cybersecurity Other

Shiraz Univ. researchers apply reinforcement learning to quantum search, shifting complexity from √D to ln(D) and achieving exponentially higher noise tolerance than Grover's algorithm. Simulations on N-qubit and D-level qudit systems confirm higher success rates—hardware vali...

AI ruling prompts warnings from US lawyers: Your chats could be used against you -

Burnout among cybersecurity leaders is through the roof. But the situation isn't just a personal tragedy for the individual. It's also a serious business risk for the organisation

www.computerweekly.com/feature/Busi...

"Start Your Security Incident Response Plan Today!" by Purnaresa Y

#cybersecurity #incident-response #security

Memo: Snap CEO Evan Spiegel says the company plans to lay off ~1,000 full-time employees, or 16% of its global workforce, to cut costs and achieve profitability (Alexandra S. Levine/Bloomberg)

Main Link | Techmeme Permalink

Gemini app starts rolling out Personal Intelligence globally | 9to5 Google

Great news for KDE Plasma users!

itsfoss.com/news/kde-pla...

Anthropic, Google, Microsoft paid AI bug bounties – quietly www.theregister.com/2026/04/15/c...

According to Ransomware.live, dragonforce ransomware group has added Curtis Design Group to its victims.

bdsmovement.net/israeli-spyw...

Why Choose CCC™ Certification for a Cybersecurity Consulting Career?

Discover the USCSI CCC™, a mid-level credential for cybersecurity consulting roles. Learn what it covers and how it advances your cybersecurity career today!

Read more: bit.ly/3O7RfWV

AI Agents Control and Trust Issues
Reliability 82% | Impact 66%
https://newshive.geekybee.net/stories/256b5940-fd24-4d1a-9cce-d46534f396a0
+1 more updated this hour.

Fractured Dream

#synthart #GeminiAi #Gemini #NanoBananaPro #aiart #AIartcommunity

CESNET and Ribbon demonstrated Layer 1 optical encryption achieving quantum security in live networks without latency overhead, moving QKD from research labs to real-world infrastructure deployment in optical networks.

#QuantumSecurity #QKD #News

📰 Turn your best AI prompts into one-click tools in Chrome

🔗 https://blog.google/products-and-platforms/products/chrome/skills-in-chrome/

💬 Discuss on HN

Anthropic Rebuilds Claude Code Desktop App Around Parallel Sessions

Sectigo is advancing post-quantum readiness by adding private PQC certificate management into its Certificate Manager platform. The upgrade helps organizations prepare for quantum threats by enabling quantum-safe cryptography and smoother migration.

#Quantum #QuantumComputing #technology #Science

今日のHuggingFaceトレンド

google/gemma-4-31B-it
このリポジトリは、Google DeepMindが開発したオープンウェイトのマルチモーダルAIモデル「Gemma 4」に関するものです。
テキストと画像を入力とし、テキストを出力します。
多様なサイズ（E2B, E4B, 26B, 31B）が用意され、テキスト生成、コーディング、推論などのタスクに適しています。
Apache 2.0ライセンスで公開されています。

Or AI content based on AI content

AI is doomed to fail in a world where quantity trumps quality of output

Further, who wants to read all this AI text produced by fast written prompts

It has been a good thing that writing took time

Best: AI use combined with the lack of critical thinking

#AI

#hacking #docker #security

#GoogleCloud highlights new contributions to #PostgreSQL, improving logical replication, upgrade processes & overall system stability while collaborating with the upstream community.

🔗 Learn more: bit.ly/3Qcz6I3

#InfoQ #AI

rewatching deep space nine and whenever the engineers have to do something with the ship's computer all i can think of is people arguing with LLM content restrictions while vibe coding

Grok’s sexual deepfakes almost got it banned from Apple’s App Store. Almost.

テストにより、クロード・ミトスの攻撃能力と限界が明らかになった

Testing reveals Claude Mythos’s offensive capabilities and limits #HelpNetSecurity (Apr 14)

www.helpnetsecurity.com/2026/04/14/c...

#AI #artificial intelligence #Google #intelligence

According to Ransomware.live, lamashtu ransomware group has added VOLTERRIES to its victims.

#TechRadar Enterprise AI governance cannot live in a prompt. So where is the safety net? https://techrad.ar/ZdQD #Pro

CISPA Display 2026: New research yearbook showcases international cutting-edge research in cybersecurity and trustworthy artificial intelligence...
weiterlesen