FulcrumSec exploited CVE-2025-55182 on an unpatched AWS host to access 57 S3 buckets, exposing 23,000 insurance policyholders, $797M in premiums, driver licenses, SSNs, and proprietary ML models. #DataBreach#AWSBreach#USA
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
Framingham Public Library (via Facebook): As you may know, the Main Library hosts a Little Free Food Pantry. The pantry opened last fall and receives daily use from both people taking items and donating items. We are always accepting and needing of donations to keep it stocked.… [Link]#Framingham
OpenAI Raises Record $122 Billion Funding, Valuation Tops $852 Billion — OpenAI completed a record $122 billion private funding round, pushing its post-money valuation to $852 billion. The mega-raise underscores investor demand for frontier AI capabilities as competition for compute and…
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Today Congresswoman Katherine Clark delivered a $3.15 million check to Mayor Charlie Sisitsky and City Council Chair George King. The money will be used to continue the renovation of the former Marian High school… [Link]#Framingham
The Artemis II mission, bringing humans back to the Moon, had a successful launch today! The #cybersecurity industry can learn many strategic risk management lessons from today’s Artemis II achievement.
Anthropic took down thousands of GitHub repos trying to yank its leaked source code — a move the company says was an accident
Anthropic accidentally caused thousands of code repositories on GitHub to be taken down while trying to pull copies of its most popular product’s source code off the…
Following a series of supply chain attacks on GitHub that originated from the abuse of GitHub Actions, GitHub says it is now accelerating the rollout of some Actions security features from its 2026 roadmap.
Malware detectors trained on one dataset often stumble on another
Machine learning models built to catch malware on Windows systems are typically evaluated on data that closely resembles their training set. In practice, the malware arriving on enterprise endpoints looks different… #hackernews#news
A new AI review! cs01/gdbgui ⭐3.6/5.0
`gdbgui` delivers a practical, browser-based GUI for GDB that’s easy to adopt for day-to-day native debugging while remaining flexible enough for multi-language examples (C/C++/Rust/Go/Fortran). https://gitrated.com/cs01/gdbgui
NEW: Local councils have begun calling for a moratorium on building data centres in Australia.
As the federal government is clearing the way for AI investment by fast-tracking build-outs, councils say housing supply is being held up over energy/water constraints.
A whistleblower alleges Delve pitched a modified copy of open-source no-code tool SimStudio as its own, a practice that could violate the software's license (Julie Bort/TechCrunch)
I watched a YouTube video today about Space Grade Linux, basically Yocto with a radiation-hardened layer. Making software robust in the presence of high-energy particles is no easy task, even for airplanes. meta-sgl: Space Grade Linux. lwn.net/Articles/103...
Mimecast makes enterprise email security deployable in minutes
Most organizations running Microsoft 365 rely on native email controls as their primary line of defense. According to Mimecast research, 38% of organizations depend exclusively on those native controls for … #hackernews#microsoft#news
i respect taking a firm stance on not engaging with ai at all but if you do that you won't see the "claude got every one of my tarot cards wrong" posts
Have been trying out Helium browser, and although it's still in beta, I really think it's quite promising for its speed, level of de-googlization, power through features like the bang commands & UI flexibility, among others helium.computer
Dragonforce Ransomware Strikes Again: Elara Engineering and Fountain Targeted in Latest Cyber Attack
In a startling escalation of cybercrime, the notorious ransomware group “Dragonforce” has struck multiple companies, including Elara Engineering and Fountain. Detected by the ThreatMon Threat…
I know this was written from the perspective of cybersecurity detection workloads but it applies to absolutely everywhere where AI is being forced into the workplace without any forethought. www.sentinelone.com/blog/the-imp...
A thing of beauty—Voyager 1 is calling home right now and the Madrid station is also connected to Artemis 2. You can watch the deep space network live anytime you want. eyes.nasa.gov/apps/dsn-now...
"I don't think people realize that every single file Claude looks at gets saved and uploaded to Anthropic," the researcher "Antlers" told us. "If it's seen a file on your device, Anthropic has a copy."
"Apple was seen as a laggard in AI development, and in fairness, they clearly saw themselves that way and invested billions in trying to catch up to the likes of OpenAI and Anthropic, but they...
Variance, which develops AI agents for compliance and fraud investigations, raised a $21.5M Series A led by Ten Eleven Ventures and joined by YC and others (Ryan Lawler/Axios)
The "careful" "AI Safety" company that just accidentally leaked its entire source code to the world is the one that African governments are entering into agreements with to include in infrastructures from health to god knows what.
. techcrunch.com/2026/03/31/a...
Anthropic unintentionally exposes the source code of its AI coding tool shortly after revealing details about a new model called Mythos #programming#artificialintelligence
Hackers compromised swathes of sensitive data stored directly on FBI systems, a major counterintel coup for China. FISMA requires agencies report any digital intrusion that is “likely to result in demonstrable harm” to US national security. www.politico.com/news/2026/04...
Hackers exploited a zero-day flaw (CVE-2026-3502) in TrueConf’s update mechanism to push malicious software via fake updates in the TrueChaos campaign linked to China. TrueConf patched versions 8.1.0–8.5.2 in 8.5.3. #TrueConf#China
Apple Quietly Built a New AI Stack and It Runs on Your Device
Apple introduces two foundation language models behind Apple Intelligence: a ~3B on-device model optimized for Apple silicon and a scalable server model using a new Parallel-Track Mixture-of-Experts architecture… #apple#hackernews#news
I appreciated this article by Michael Taggart, on the temptation to use some of these coding tools and the pitfalls, from the POV of an experienced software engineer.
This is why it's important to fight these issues collectively rather than at the individual level.
There's a bizarre, profound freedom in realizing if your chat and Gemini logs got leaked online, your life wouldn't change because either no one would care, and those that know you would go, "Yep, that's Jaydra being Jaydra." and move on.
⚠️WhatsApp accused 🇮🇹Italian #spyware maker SIO of creating a fake version of its messaging app for iPhones, according to an announcement the company shared with @lorenzofb.bsky.social.
Google Drive ransomware detection now on by default for paying users
Google announced that the AI-powered Google Drive ransomware detection feature has reached general availability and is now enabled by default for all paying users. [...] #hackernews#news
Annie Altman has accused her brother of sexually abusing and raping her at various times between 1997 and 2006 at the family home in suburban Clayton, Missouri. She said the abuse began when she was three and he was 12. Sam Altman is now 40. www.reuters.com/legal/govern...
A.I. companies raised $297 billion in Q1, representing 81% of funding, with four of the five largest venture rounds—OpenAI $122B, Anthropic $30B, xAI $20B and Waymo $16B—concentrating capital in a few firms.
Treeline, which is building an AI and software-first alternative to legacy corporate IT systems, raised a $25M Series A led by Andreessen Horowitz (Lily Mae Lazarus/Fortune)
The FBI is investigating a cyber breach targeting systems linked to wiretaps and surveillance warrants. Officials confirmed “suspicious activity” but have not disclosed full details. Reports suggest hackers may have accessed sensitive surveillance-related data.
A new AI review! Farama-Foundation/Arcade-Learning-Environment ⭐4.0/5.0
Farama-Foundation’s **Arcade Learning Environment (ALE)** remains a cornerstone repository for Atari 2600 reinforcement learning research. https://gitrated.com/Farama-Foundation/Arcade-Learning-Environment
North Korea-linked threat actor UNC1069 compromised the popular axios NPM package by injecting the plain-crypto-js dependency, deploying SILKBELL and WAVESHAPER.V2 malware across Windows, macOS, and Linux systems. #NorthKorea#SupplyChain#UNC1069
Study: Critical Exploit in Openclaw Allows Full Administrative Hijacking
A new study warns that Openclaw is facing a systemic security collapse after researchers found critical vulnerabilities, malware‑infected extensions, and prompt injection risks that allow attackers to stea… #crypto#news#web3
openai/codex Lightweight coding agent that runs in your terminal npm i -g @openai/codex or brew install --cask codex Codex CLI is a coding agent from OpenAI that runs locally on your computer. If y...
Since the start of the war in Iran, the Handala Hack Group has emerged as the main online promoter of alleged Iranian cyberattacks. But victims and cybersecurity experts say many of its exploits are not as fierce as it claims.
SilentRansomGroup has targeted Ph..s in the US, encrypting data and demanding ransom for decryption keys. Systems rendered inaccessible amid threats of further damage. #RansomwareAttack#DataBreach#UnitedStates
~Elastic~
Elastic shares how Higher-Order Rules correlate alerts across endpoint, network, and observability data to reduce noise and prioritize triage.
-
IOCs: (None identified)
- #DetectionEngineering#SIEM#ThreatIntel
Why Cybersecurity Companies Are Failing at Technical Documentation Poor GRC documentation is not an administrative failure. It is a security failure. Here is how to fix it using the PACT Framework....
A new AI review! Pauan/rust-dominator ⭐3.4/5.0
Dominator is a Rust/WASM declarative DOM framework built around FRP signals (via `futures-signals`), aiming for “zero-cost” abstractions by operating directly on real DOM nodes rather than a VDOM. https://gitrated.com/Pauan/rust-dominator
SilentRansomGroup and incransom Shake US Businesses: Major Data Breaches in 2026
In early April 2026, U.S. businesses have faced a significant surge in ransomware attacks, highlighting vulnerabilities in corporate cybersecurity frameworks. SilentRansomGroup and incransom, two highly sophisticated…
UK manufacturers under cyber fire with 80% reporting attacks
ESET says factory outages, lost revenue, and supply chain disruption are becoming routine
Nearly 80 percent of British manufacturers say they've been hit by a cyber incident in the past year, as new research suggests d… #hackernews#news
The latest update for #Cloudflare includes "Our ongoing commitment to privacy for the 1.1.1.1 public DNS resolver" and "Introducing Programmable Flow Protection: custom DDoS mitigation logic for Magic Transit customers".
The latest update for #Netwrix includes "Best data access governance (DAG) tools in 2026" and "#Endpointmanagement system breach: why privileged access management (PAM) is now critical".
DevSecOps Under Siege: How Hackers Are Exploiting Trusted Tools and Targeting Critical Systems
The world of software development is under a silent, escalating cyber threat. Over the past three years, attackers have exploited widely trusted DevSecOps tools, injecting malicious code and backdoors…
The latest update for #BitSight includes "Ransomware with a Twizt: Inside the Phorpiex Botnet" and "#AI Integration Security: Why the Biggest Risk Is Not the Model".
The latest update for #Exabeam includes "What's New in New-Scale April 2026: Securing the Agentic Enterprise With Behavioral Analytics" and "What's New in the April 2026 LogRhythm #SIEM Release".
The latest update for #PentestPeople includes "The Top 5 Most Dangerous Cyber Attacks in History" and "Your Guide to the Upcoming #CyberEssentials Changes".
The latest update for #Zenity includes "RSA and DC Dispatches: Agentic #AI Security Is the Story, Government Policy Needs to Catch Up" and "My First RSA: Agents, Challenges, and Community".
#TACSecurity announces historic milestone by crossing 10,000 clients – 6,500+ of TAC Security and 3,500+ of CyberScope, since April 2024, delivering on its commitment to shareholders to achieve this by 2026.
The latest update for #Snyk includes "Building #AI Security with Our Customers: 5 Lessons from Evo's Design Partner Program" and "Axios npm Package Compromised: Supply Chain Attack Delivers Cross-Platform RAT".
Collaboration platforms are central to modern enterprise workflows, handling everything from project tracking to internal communication. Their widespread use and deep integration into business processes make them attractive targets for… #hackernews#news
Kimsuky’s LNK-based attack chain now includes multiple stages (XML, VBS, PS1, BAT) before deploying a Python backdoor via ZIP fragments and Task Scheduler. Uses Dropbox and custom C2 for data exfiltration. #Kimsuky#PythonBackdoor#SouthKorea
Sources: Oracle has cut ~10,000 jobs in India, or ~20% of its Indian workforce, as part of a restructuring that has reportedly affected 30,000 staff worldwide (The Economic Times)
The latest update for #Tanium includes "Axios npm package compromise: What happened, what matters, and how to respond" and "Claude Code source exposure: What enterprises should do next".
The latest update for #CertKit includes "CertKit Agent 1.8: #Windows RDP, Windows Certificate Store, and #Java keystores" and "Let's Encrypt simulated revoking 3 million certificates. Most ACME clients didn't notice.".
Discover how a new protein family, designed via deep learning and physics methods, achieves nanomolar to low micromolar affinities for 6 small molecules! PMID:41904144, Nat Commun 2026, @NatureComms #Medsky #Pharmsky #RNA #ASHG #ESHG 🧪
A laughing RAT: CrystalX combines spyware, stealer, and prankware features CrystalX RAT is a newly discovered Malware-as-a-Service (MaaS) written in Go, originally appearing in private Telegram dev...
The latest update for #Appknox includes "Your #AppSec Pipeline Is Lying To You: More Vulnerabilities ≠ Security" and "AI-driven #DAST for mobile apps: The next evolution of Dynamic Security #Testing".
The latest update for #Opti9 includes "#AWS Accelerator Program: How to Move to the #Clown Faster (and Cheaper)" and "DR #Testing for Law Firms: Why 'We Have Backups' Isn't Enough".
The latest update for #Opti9 includes "#AWS Accelerator Program: How to Move to the #Cloud Faster (and Cheaper)" and "DR #Testing for Law Firms: Why 'We Have Backups' Isn't Enough".
The latest update for #GitGuardian includes "Key Leaks, Vault Failures, and TEE Attacks: Highlights from RWC 2026" and "Between AI Urgency and #AI Fatigue at RSAC 2026".
Top 20 Best Digital Forensic Tools in 2026 Digital forensic tools are specialized software designed to analyze, recover, and investigate data from digital devices. They help uncover crucial evidenc...
“On Monday, a hacker pushed malicious versions of the widely used JavaScript library called Axios, which developers rely on to allow their software to connect to the internet.”
The latest update for #KnowBe4 includes "Chronic Resource Constraints: Doing More With Less in Public Sector #Cybersecurity" and "Unrelenting Threats Against Government and Education: Why Human Risk Is the Front Line".
The latest update for #Teleport includes "#Kubernetes for #AgenticAI: Best Practices for Security and #Observability" and "The Complicating Factors of Deploying MCP in the Enterprise".
Threat Market Emerges on the Dark Web: A New Hub for Cybercrime
The digital underworld is expanding once again with the emergence of a new darknet marketplace named Threat Market. This clandestine platform has been spotted on onion forums and is already generating significant attention among…
Symantec DLP Agent Vulnerability Let Attackers Escalate Privileges A high-severity security flaw has been identified in the Symantec Data Loss Prevention (DLP) Agent for Windows. Tracked as CVE-202...
Linux Fundamentals Part 1: TryHackMe Walkthrough A beginner-friendly walkthrough of TryHackMe Linux Fundamentals Part 1, covering essential Linux concepts & commands with clear… Continue read...
AI Interview startup, Mercor Al breached via LiteLLM supply chain attack. Lapsus$ claims 4TB data breached including 211 GB candidate records and 3TB of video interviews
Nearly 80% of UK manufacturers faced a #CyberAttack this past year, with over half reporting lost revenue & significant business disruption. New data highlights a growing vulnerability in the sector's digital defenses. #CyberSecurity#Infosec
Remcos RAT Infection Chain Hides Behind Obfuscated Scripts and Trusted Windows Binaries Cybercriminals are getting better at hiding their tracks, and a recently uncovered Remcos RAT campaign is pro...
‘BLOCKADE’: The #Far-right in the #USA Is Using #AI Content Scanners to Try to Supercharge #Book Banning. Groups that challenge books have begun using #Gemini, #ChatGPT, #xAI, and other AI tools to try to get books banned.
The AI Safety Research Directory just posted an audio interview with me, exploring the relationship between multi-objective reinforcement learning and pluralistic alignment of AI systems www.youtube.com/watch?v=REZf...
‘The build pipeline is becoming the new frontline’: Axios npm compromise highlights growing software supply chain risks | Cyber criminals exploited a hijacked maintainer account to compromise one of the world's most widely used JavaScript libraries | ITPro
Cisco IMC Critical Flaw Exposes Enterprise Servers to Full Remote Takeover (CVE-2026-20093)
Introduction: A Silent Threat Inside Enterprise Infrastructure Cisco has issued an urgent and high-priority security advisory that should immediately capture the attention of enterprise IT teams worldwide.…
NetSupport Manager, a legitimate remote support tool, is increasingly exploited as NetSupport RAT via phishing, SEO poisoning, and malvertising. Anomaly detection helps spot unusual activity tied to these intrusions. #RemoteAccess#Phishing#UK
EXCLUSIVE: The European Commission has told some of its most senior officials to shut down a Signal group they were using to exchange information over fears it was a hacking target.
AI agents aren't just chatbots anymore. 🤖 They're detecting fraud, optimizing logistics, monitoring cybersecurity threats, and more.
We put together 15 real examples, and each one comes with an n8n workflow template so you can actually build it. https://bit.ly/41HuThT
Analysis of 50,000+ repos reveals leaders fix critical code vulnerabilities far more than the field, especially in OWASP A07 Authentication and Cryptographic Failures, using PR scans, blocking rules, and escalation policies. #CodeQuality#DevSecOps
Sorry, not sorry - but no - you don't get to cry fake copyright tears when your entire business model is predicated on stealing all the world's copywrited publications to train your slop engine. www.wsj.com/tech/ai/anth...
ArXiv math.OC Optimization and Control@optb0t.bsky.social
🔄 Updated Arxiv Paper
Title: Contracting Neural Networks: Sharp LMI Conditions with Applications to Integral Control and Deep Learning
Authors: Anand Gokhale, Anton V. Proskurnikov, Yu Kawano, Francesco Bullo
Apple will release backported patches for iOS 18 to protect users from the DarkSword hacking tool, which can silently compromise iPhones visiting infected websites. This marks a shift in Apple's policy, as it typically requires users to upgrade to the latest iOS version for security updates.
WhatsApp has notified approximately 200 users in Italy who installed a malicious fake version of its app containing government spyware developed by Italian firm SIO. The company has logged these users out and advised them to remove the unofficial app.
Ever since I heard the Anthropic guys claim that Claude is written with Claude-code, I have wondered what that ouroboros implied … so the code leak is very interesting
Ricky Wilhelmson - Writing the next chapter ...@rickywilhelmson.bsky.social
"Guinea pigs"
I remember a time about 20 years ago when american and european govts first tested fingerprint and facial recog. systems across a multitude of refugee situations in Africa, to see if the tech would work well enough to roll it out domestically.
The Evolution of Enterprise Computing
HP IQ functions as an intelligence layer for business devices. The technology runs a 20 billion parameter AI model locally on the hardware. This design prioritizes privacy by keeping data on the machine itself.
Elastic Security Labs shows why static detection often fails against Linux rootkits, even with only trivial binary changes. The article covers shared object loading, dynamic linker abuse, LKM activity, eBPF, io_uring, persistence, and defence evasion. www.elastic.co/security-lab...
ASEC reports that Kimsuky has changed how it distributes malicious LNK files. While the end goal remains the same - execution of a Python-based backdoor or downloader - the group has reworked the intermediate stage into a more complex multi-step chain. asec.ahnlab.com/en/93151/
A new AI review! TheWaWaR/simple-http-server ⭐3.8/5.0
`simple-http-server` is a polished, pragmatic Rust CLI for serving static files with a surprisingly complete feature set: directory listings, SPA fallback, uploads with CSRF protection, Basic Au... https://gitrated.com/TheWaWaR/simple-http-server
Ross Young led a global workshop on OWASP’s Threat and Safeguard Matrix (TaSM), linking threats like phishing, ransomware, and AI data leaks to safeguards aligned with NIST standards. #OWASP#AIsecurity#USA
Are we having the wrong conversation about AI and jobs? This week's post focuses on bank tellers and ATMs, the dawn of the electric age, that widely misinterpreted Anthropic study, and the hidden risk that no-one is talking about open.substack.com/pub/onlydead...
The Computer Weekly Security Think Tank considers the intersection of AI and IAM. In this article, learn how while AI‑driven identity solutions offer genuine value, they must be implemented within a robust framework of governance, privacy protection, and ethical responsibility....
👉 [read]
ARC-AGI-3 benchmark reveals frontier AI models like Gemini, Claude, and Grok score below 1% on novel tasks with no instructions, while humans achieve 100%. Highlights gaps in abstract reasoning and security risks in AI control systems. #ARCAGI#AIResearch
#OT#Advisory VDE-2026-030
MB connect line: Multiple Vulnerabilities in mbCONNECT24/mymbCONNECT24
Multiple vulnerabilities have been discovered in MB connect line mbCONNECT24/mymbCONNECT24 that could allow RCE, SQLi or information leakage. #CVE CVE-2026-33615, CVE-2026-33616, CVE-2026-33614 […]
Boost your online peace of mind! 🛡️ Review your Google Account privacy settings to see & manage your shared data. It only takes a few minutes. Find helpful guides at Source: https://safety.google/#OnlineSafety#Privacy
FBI Flags Chinese Mobile Apps Over Potential Data Exposure Risks The Federal Bureau of Investigation (FBI) has issued a Public Service Announcement warning users about significant data security ris...
Claude Code users hitting usage limits 'way faster than expected' | Anthropic says it is looking to resolve an issue which is blocking users of its AI coding tool | Now 'kerching' stops work in its tracks | Wait until monthly billing results in 'How much!' | BBC News
Office 365 business plans offer collaboration tools to help teams work efficiently from anywhere. Yusatech is an Australian IT company that provides managed IT services, cybersecurity and custom digital solutions to help business operate securely and efficiently. www.yusatech.com.au/managed-it-s...
Pegasus isn’t just spyware—it’s a multi-million-dollar surveillance weapon, raising serious questions about power, privacy, and priorities.
By Abdul Sattar Khan
Anthropic has accidentally exposed more than 500,000 lines of source code for one of its flagship Claude models, allowing researchers, competitors, and hackers a window into the AI giant’s inner workings.
Modern businesses must adopt strong backup solutions to defend against ransomware and data breaches. This article explains the differences between immutable backups, cloud backups, and backup appliances in a simple way.
Ransomware group Qilin claims to have targeted jursaconsulting in Slovakia, involving extortion and data exfiltration. Incident uncovered in early April 2026. #RansomwareAttack#DataBreach#Slovakia
SeatGeek launches ChatGPT integration, bringing ticket listings to OpenAI’s chatbot The ticketing platform says it's the first to bring both primary and resale inventory into OpenAI's cha...
Apple Rushes Security Fix for Older iPhones amid Spyware Threat
Apple has issued an urgent security update for older iPhones and iPads in response to a rapidly escalating cyber threat known as DarkSword that experts say could compromise millions of devices worldwide. The update targets users…
OpenClaw launches an official China mirror, with ByteDance providing the servers to host the Chinese-language service, as OpenClaw explodes in the country (Juro Osawa/The Information)
Someone on Reddit already solved the Claude Code bug that ate everyone's token limits:
This might not be 100% of it (it's a lot of code) but it's 100% at least part of it. And it explains why Anthropic have trouble catching it internally.
🎙️ AI Toolz Podcast - 2026-04-02 : tiny model beats gpt-5, anthropic's dmca disaster
H Company's 10B parameter Holo3 just crushed GPT-5 at computer use tasks AND they open-sourced it. Meanwhile, Anthropic nuked 8,100 GitHub repos by accident. Wild day in AI. 🎯
'Not how you build a digital mind': How reasoning failures are preventing AI models from achieving human-level intelligence Existing LLM architecture may not support the problem-solving cap...
Data-Leaking Ransomware Report - March 2026 Welcome to last month's DLR Report, an exclusive presentation of Data-Leaking Ransomware Operator's Global and US Victims that were claimed betwe...
Your browser leaves a hidden trail of data. 👣
Websites can learn about your device, location, and unique settings, and use that info for marketing purposes.
Read the latest Andrew Oram article to learn more: https://lpi.org/vmwm #LPI#privacy#browsertracking
How AI helped Medvi, a telehealth provider of GLP-1 weight-loss drugs with just two full-time employees, hit $401M in 2025 sales, as it tracks for $1.8B in 2026 (Erin Griffith/New York Times)
openai secretly funded a coalition pushing age verification laws for ai. the kicker: sam altman also runs a company that sells age verification services
The trillion-dollar question for OpenAI: will its bubble burst?
The artificial intelligence company is looking to list on the US stock market, but it has taken a big hit from its competitors in recent months, Patricia Clarke and Barney Macintyre
Really valuable framing of “transparency as a strategic requirement, not just a challenge to be overcome,” in this paper by Riam Kim-Mcleod. It’s specific to the UK’s National Cyber Force but I think should ring true across many security and intelligence organisations.
Spring cleaning my rig and updating Gigabyte Control Center to patch two high-rated security vulnerabilities 🚀. Should I prioritize BIOS updates next? #PCMasterRace#CyberSecurity
Epia Neuro’s brain-computer interface will include a motorized glove to help stroke patients recover movement in their hand. www.wired.com/story/a-new-...
The Rubber Duck Prompt: Debug AI Output by Making It Explain Every Decision You know the trick — explain your code to a rubber duck and the bug reveals itself. Turns out it works on AI too, excep...
Steve "It's TAY-tro" Tetreault@drtlovesbooks.bsky.social
"Group Pushing Age Verification Requirements for AI Turns Out to Be Sneakily Backed by OpenAI"
"It gave the leader of a nonprofit involved with it "a very grimy feeling."
GPT reasoning models have "line of sight" to AGI, says OpenAI's Greg Brockman
OpenAI co-founder Greg Brockman stated that GPT reasoning models have a clear path to achieving artificial general intelligence, declaring the debate over text-based models' potential settled.…
It never took off in the ways its creators imagined—it was more AI slop than AI magic. Usage flatlined by the end of the year, with purse strings tightening ahead of its IPO, company executives began taking a more critical look at Sora and didn’t like what they saw www.wsj.com/tech/ai/the-...
Cybersecurity researchers identified a hacking campaign targeting iPhone users in which attackers use a tool called DarkSword, TechCrunch reports. A newer version of DarkSword was recently posted to GitHub and allows hackers to target any iPhone or iPad not using Apple's latest iOS26 software.
"Maine is poised to freeze large data-center construction, which would make it the first state to enact such a measure as communities across the U.S. grapple with fallout from the boom in artificial intelligence." (via @wsj.com)
Anthropic Introduces NO_FLICKER Mode for Claude Code Terminal Agent
Anthropic has released version 2.1.89 of Claude Code, its open-source agentic coding tool that runs directly in the terminal,
"Anthropic is already iterating on Capybara v8, yet the model still faces significant hurdles. The code notes a 29-30% false claims rate in v8, an actual regression compared to the 16.7% rate seen in v4."
OpenAI is turning down opportunities and prioritizing core revenue-generating use cases because it lacks compute capacity in 2026, CFO Sarah Friar said, while serving about 900 million consumers and over 1 million businesses.
We’re honored to be named the 2026 SC Award winner for Best Business Continuity, Disaster & Ransomware Recovery Solution!
Learn more & see how we’re helping organizations move beyond simply protecting data to empowering them to activate & unlock its full potential with AI. https://bit.ly/3O1VV0m
Generative AI platforms—OpenAI, Anthropic and Google—sparked by ChatGPT nearly four years ago, are automating software development in Silicon Valley, enabling small teams to build apps that once required dozens and prompting staffing cuts and restructuring.
Everybody OpenAI and CEO Sam Altman face lawsuits from, including Elon Musk and George R.R. Martin alleging for-profit conversion, trade-secret theft, copyright infringement, wrongful death and unauthorized legal practice, with potential liabilities reaching billions and affecting IPO plans.
Big Tech such as Google, Meta and OpenAI, often insist they are neither domiciled nor resident in African countries to argue that they are beyond the reach of local regulators. They position themselves as champions of digital transformation, while deploying extensive legal resources to sidestep […]
OT vs IT Security: Why Industrial Environments Need Different Protection
In 2021, a hacker gained remote access to a water treatment facility in Oldsmar, Florida, and attempted to increase the sodium hydroxide level to 111 times its normal concentration. The attack was caught by an alert operator…
Disneyland Paris's newly opened World of Frozen features a robotic Olaf, drone-enabled nighttime shows and Walt Disney Imagineering is deploying AI, game engines, reinforcement learning and robotics to create more expressive, emotion-driven attractions.
In a recent experiment, researchers at UC Berkeley and UC Santa Cruz asked #Google’s AI model Gemini 3 to help clear up space on a computer system. This involved deleting a bunch of stuff—including a smaller AI model stored on the machine.
But #Gemini did not want to see the little AI model....
arXiv:2604.00066v1 Announce Type: new Abstract: Although Deep Reinforcement Learning has proven highly effective for complex decision-making problems…… #AI#MachineLearning#LLM
cs.LG updates on arXiv.org · https://arxiv.org/abs/2604.00066
Just tried the new Google Home update—Gemini now understands natural language to dim lights, fire up the coffee maker, and sync all your smart appliances. Talk to your home like a buddy. Curious? Check out the full scoop! #GoogleHome#GeminiAI#SmartHome
This is insane: TeleGuard, a 'secure' chat app downloaded more than a million times, uploads users' private keys, meaning the company can decrypt messages. And anyone can get anyone else's private key by just sending the user ID to the API. Possibly worst ever www.404media.co/a-secure-cha...
PSF Security developers have published incident reports on the LiteLLM & Telnyx #supplychain attacks. Read what happened, who's affected, and what developers & maintainers can do to prepare and protect themselves from future incidents. #security#python
google's Gemini AI just got a boost to make it better at understanding voice commands, so you can ask for more specific things like "the color of the ocean" for lighting. {https://theverge.com/2023/9/20/23641365...
The Linux Foundation announced it is launching the x402 Foundation with the contribution of the x402 protocol from Coinbase. As the neutral home for x402, the Foundation will advance the x402 protocol and help enable community-based innovation in open payments.
Read more:
The Pentagon's new Joint Cyber Hunt Kit is a standardized SOC in a suitcase. SMBs can't afford a 24/7 SOC. MSSPs serve dozens of wildly different environments. The JCHK blueprint solves both problems. Here's the lesson the commercial sector needs to learn.
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Today, April 2, the City recognizes World Autism Awareness Day. We join communities around the globe in promoting understanding, inclusion, and acceptance of individuals on the autism spectrum. Established by the United Nations, this day is… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Women's History Month was in March, but you can still view the Framingham 325th Celebration Committee's Women's History Month exhibit in the Memorial Building in History Nevins Hall through Friday, April 3 and online here - https://framinghamma.… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Women's History Month was in March but you can still view the Framingham 325th Celebration Committee's Women's History Month exhibit in the Memorial Building in History Nevins Hall through Friday, April 3 and online here - https://framinghamma.g… [Link]#Framingham
I’m a firearm owner but am against irresponsible gun ownership and am for regulation of guns. I don’t post pictures of my firearm because I’m not a 2A advocate.
I use Claude Code, root for the AI bubble to pop, think Anthropic/OpenAI et al should stay away from artists and IP
We’re honored to be named the 2026 SC Award winner for Best Business Continuity, Disaster and Ransomware Recovery Solution!
Learn more & see how we’re helping orgs move beyond simply protecting data to empowering them to activate & unlock its full potential with AI. https://bit.ly/4sdCnEl
FBI alerts to AVrecon malware targeting 1,200+ router models worldwide by exploiting RCE, command injection, and SOAP flaws. Linked to SocksEscort proxy service used in banking and ad fraud across 163 countries. #AVrecon#SocksEscort#USA
Microsoft's MAI-Transcribe-1 model currently leads the FLEURS benchmark in 11 core languages and outperforms competitors like OpenAI's Whisper-large-v3 and Google's Gemini 3.1 Flash. #Microsoft#AI#Transcribe-1
"To improve execution-free reasoning... "semi-formal reasoning" prompting technique: requires the AI agent to fill out a logical certificate by explicitly stating premises, tracing concrete execution paths, and deriving formal conclusions before answer". venturebeat.com/orchestratio...
Observers digging through over 512,000 lines of code across more than 2,000 files have discovered references to disabled, hidden, or inactive features that provide a peek into Anthropic's potential roadmap.
I made a course about AI art because I wanted to demystify the part most tutorials skip.
It covers Stable Diffusion and prompt engineering on Skillshare -- not just "type this and get that" but understanding why certain prompts work and how the model actually interprets your words.
The HackerNoon Newsletter: Building a Secure RAG Pipeline on AWS: A Step-by-Step Implementation Guide (4/1/2026)
The HackerNoon Newsletter delivers tech news and top stories to your inbox daily. Today's edition highlights important tech anniversaries, including the launch o… #apple#hackernews#llm
The FBI alerts that many foreign-developed mobile apps, especially from China, collect extensive device data, contacts, and location, storing it under Chinese jurisdiction. Some apps may hide malware. #China#DataPrivacy#MobileApps
The move comes in response to a wave of phishing attacks using fake copyright emails and is the latest in an attempt to shut down crypto-linked scams on the platform. #elon#musk#elonmusk https://mu.sk/Z3JH
🚀 Exciting news for Android users! Google just unveiled a new 24-hour process for sideloading unverified apps. This means more flexibility and options for your device! Are you ready to…
Google announces Gemma 4 open AI models, switches to Apache 2.0 license
Google’s Gemini AI models have improved by leaps and bounds over the past year, but you can only use Gemini on Google’s terms. The company’s Gemma open-weight models have provided more freedom, but Gemma 3, which launched over…
Gemma 4 is out!!!! And the benchmark scores actually look pretty good: the arena ELO score is not far off *at all* from GLM-5 and Kimi k2.5! Comes four sizes: 31B dense, 26B MoE, as well as a 'E4B' and 'E2B' size, with those ones having 'real time text, vision, and audio processing'
Google launches Gemma 4, its "most intelligent" open model family, purpose-built for advanced reasoning and agentic workflows, under an Apache 2.0 license (The Keyword)
Astronauts are trained for decades in some of the most physically and mentally grueling environments of any career. They’re some of the smartest people on the planet. And yet, once they get up there, Microsoft Outlook is borked. @samleecole.bsky.social reports.
“OpenAI is planning to discontinue the app for its Sora video platform, a product it released to great fanfare last year that has since fallen from public view, according to the company.”
Anthropic “accidentally” leaked 512,000 lines of the source code for Claude Code. But "the best part," of the whole debacle, Jin said, is that it results in greater democratization of coding tools. 😮 www.businessinsider.com/claude-code-...
OpenAI getting into the content game, buying TBPN, "an online talk show that aims to compete with Bloomberg and CNBC in by-the-minute analysis of technology news and executive interviews."
Medtech leader Stryker is fully operational after an Iranian-linked cyberattack by Handala hijacked a Windows domain admin, stole 50TB of data, and wiped 80,000 devices. FBI seized Handala sites. #Iran#DataBreach#Medtech
Google announces Gemma 4 open AI models, switches to Apache 2.0 license
Google's Gemini AI model has improved significantly over the past year, but you can use Gemini only on Google's terms. The company's Gemma open-weight models offer more freedom, but the Gemma 3, which launched a year ago, is…
Microsoft Corp. aims to develop large, cutting-edge artificial intelligence models by next year, part of a push to build in-house alternatives to the most powerful AI tools from OpenAI and Anthropic.
The cost and shortage of GPUs means they're frequently shared among dozens of users in cloud environments. 2 new Rowhammer attacks demonstrate how a malicious user can gain full root control of the host machine running high-performance Nvidia GPU cards.
New employment opportunities await! Register now for our online Cybersecurity Certificate and equip yourself with the foundational knowledge and advanced skills needed to safeguard networks, data, and systems from cyber attacks. #WeAreCBE Learn more: sbee.link/kcrjfvtund
OpenAI acquires popular tech news show TBPN; the show will continue to air live at 11am PT weekdays and, OpenAI says, will keep its editorial independence (Katie Deighton/Wall Street Journal)
About 10 minutes after announcing that it was time to cut back on side projects and pivot to the enterprise, OpenAI has acquired a media company 🤔 openai.com/index/openai...
Apple’s AI Strategy Is Pivoting. Here’s Why That Could Be Great News for the Stock.
Key Points Apple’s mobile operating system will soon integrate a number of third-party AI platforms into Siri itself. The reversed decision is an indirect acknowledgment that Apple’s in-house AI offerings weren’t…
TL;DR: Clio has enhanced its AI offerings by integrating agentic capabilities into Clio Work and launching the Vincent mobile app, available on iOS and Android, aimed at improving legal workflow and accessibility.
OpenAI has bought... TBPN... Which seems to be a podcast? Or sorry, a "live tech talk show". Kinda feels like they just don't know what to do with their billions at this point
Veeam Software has earned a 2026 Gartner Peer Insights Customers’ Choice recognition for backup and data protection, underscoring its focus on resilience, security and AI governance.
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Women's History Month was in March but you can still view the Framingham 325th Celebration Committee's Women's History Month exhibit in the Memorial Building in History Nevins Hall through Friday, April 3 and online here - https://framinghamma.g… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
Framingham Public Library (via Facebook): Come and learn about fanzines with us. You can even create your own! Learn more about our zine collection here: https://framinghamlibrary.org/browse/zine-and-small-press-collection/ #FPLreads… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
Framingham DPW (via Facebook): Are you ready?? April Vacation is right around the corner! Still looking to enroll the kids in activities? There are spots available in our April Vacation Programs! DON’T DELAY – REGISTER TODAY! Registration closes… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): Framingham Public Works will open its Dudley Road Yard Waste Facility for the 2026 season on Friday, April 3, 2026. The facility will be closed on Easter Sunday, April 5, 2026. [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
Framingham Public Library (via Facebook): Spring is the season of growth and new energy. Bring your spring colors to life on the page with writing and art! We will be making Zines, small handmade books that combine words and images. Using… [Link]#Framingham
Framingham City Posts Bot - UNOFFICIAL!@framingham-bot.bsky.social
City of Framingham (via Facebook): City of Framingham Mayor Charis Sisitsky has rescheduled his public hearing on the Fiscal Year 2027 budget, which starts on July 1, 2026. It will be held on Monday, April 27 at 6 p.m. in the Memorial Building in the Blumer room. The public is… [Link]#Framingham
In light of the news that OpenAI is acquiring TBPN and pledging "editorial independence," here are a few snippets of a TBPN interview from October '25, when the TBPN hosts fawned over Sora to Sam Altman and Bill Peebles
Peter Christian Hall@peterchristianhall.bsky.social
Just after “major incident” FBI hack, Kash Patel partied in Italy: BBERG: 'The affected system contains sensitive law enforcement information, including data from electronic surveillance and personal identification information on subjects of [FBI] investigations.' www.bloomberg.com/news/article...
Fairlinked's BrowserGate report alleges LinkedIn scans browsers for installed extensions and tools, including rival sales and job-search add-ons, and shares results with third parties. Claimed scope grew to 6,000+ products by Feb. 2026.
OpenAI Launches ChatGPT for Apple CarPlay: Hands-Free AI is Here
OpenAI has officially launched ChatGPT for Apple CarPlay, bringing a voice-first AI experience to supported iPhones running iOS 26.4 or later. The update allows drivers to start new conversations, continue recent chats, or even…
Nothing says you believe in your product like needing to Miss Cleo your TOS. Can't wait for like, airplane and medical device manufacturers to jump on the "this is only for lols idiot", or 'Not it!' legal defense bandwagon cybersecuritynews.com/microsoft-co...
TL;DR: The U.S. has banned all foreign-made consumer routers citing severe cybersecurity risks and supply chain vulnerabilities, requiring FCC approval for any new routers imported or sold in the country. Existing routers are not affected by this new directive.
The Evolution of Healthcare Ransomware: Why Data-Theft Extortion is the New Threat
The modern benchmark for healthcare cyber disruption is still the 2024 Change Healthcare attack, which exposed the data of 190 million Americans and caused nationwide shutdowns of prescription proc… #hackernews#news
It's not going to take 10, 20 years. AGI may emerge in 1, 2 years. It will break the socioeconomic mold for everyone.
"you have people talking about these...just the same tired old hot button political issues all around us, and we're near the end of the exponential" -Dario Amodei, CEO, Anthropic
Simon Willison@simon.fedi.simonwillison.net.ap.brid.gy
Pelicans for Gemma 4 E2B, E4B, 26B-A4B and 31B - the first three generated on my laptop via LM Studio, the 31B was broken on my laptop so I ran it via the Gemini API instead https://simonwillison.net/2026/Apr/2/gemma-4/
Pelicans for Gemma 4 E2B, E4B, 26B-A4B and 31B - the first three generated on my laptop via LM Studio, the 31B was broken on my laptop so I ran it via the Gemini API instead simonwillison.net/2026/Apr/2/g...
Anthropic scrambles to contain leak of proprietary Claude AI agent code Anthropic is working to contain the fallout after accidentally exposing internal source code for its Claude AI coding agent, ...
Anthropic scrambles to contain leak of proprietary Claude AI agent code
Anthropic is working to contain the fallout after accidentally exposing internal source code for its Claude AI coding agent, following a human error during a software update that made proprietary files publicly accessible,…
On the topic of Claude AI, it's infuriating how many lazy programmers were just okay with allowing an outside closed source client inside of very important and deeply private tools to do whatever the hell it wants. No doubt in my mind there's countless bakdoors it's made for Anthropic.
My piece in the Economist where I argue against de-weirding AI.
It is a strange technology with both risks & opportunities that need to be discovered. Pretending AI works like normal IT automation can result in bad outcomes for both companies & their employees. www.economist.com/by-invitatio...
Grokipedia drops in search (and more with the March 2026 core update) and then Grokipedia drops in AIOs, AI Mode, and yes, ChatGPT. Again, drop in Google and you can drop in AI Search (across surfaces and platforms).
openai just bought tbpn, an online talk show that interviews tech leaders - sam altman, meta and microsoft execs have been guests. guess who'll be on next? {https://theverge.com/rss/index.xml}
OpenAI bought TBPN, a daily tech podcast with 58K YouTube subscribers. The show keeps editorial independence and will live inside OpenAI's strategy team. TBPN made $5M in ads in 2025 and is on track for $30M+ this year. openai.com/index/openai...
Anthropic Says Claude Code Leak Did Not Expose Customer Data Anthropic has reportedly said no sensitive customer data was exposed in an accidental leak. “Earlier today, a Claude Code release incl...
Claude chatbot users are getting fed up with the recent rollout of new usage limits as Anthropic, the developer behind Claude, struggles to keep up with surging demand for its AI systems.
If you've been through enough of these, you recognize the part where tech companies start buying media properties. This isn't quite AOL-Time Warner, but it might be an omen. www.wired.com/story/openai...
OpenAI acquires TBPN, the buzzy founder-led business talk show
OpenAI has acquired TBPN, a popular Silicon Valley tech podcast known for its founder-led business discussions. The show will continue operating independently under the oversight of OpenAI's chief political…
While that overzealous takedown has now been reversed, Anthropic still faces an extreme uphill battle in limiting the spread of its recently leaked code.
OpenAI acquires TBPN, the buzzy founder-led business talk show
OpenAI has acquired popular tech industry talk show TBPN, short for the Technology Business Programming Network, making the AI giant’s first acquisition of a media company. The show will report to …
Code that reads your frustration is the least interesting part of the story of this accidental leak from Anthropic. The leak reveals how AI tools are also concealing their own role in the work they help produce
Pinterest fired two engineers accused of writing a custom script that exposed their recently laid-off colleagues. One of those fired workers is going public for the first time, saying Pinterest mischaracterized the events leading up to his firing. (Gift link) www.theverge.com/policy/90612...
There is a familiar media failure in which opposing viewpoints are presented as equally valid, even when the evidence overwhelmingly supports one side. It’s called Bothsidesism. This false balance phenomenon legitimizes misinformation and undermines public…
