A new survey found 54% of enterprises have already experienced an AI agent security incident or near-miss, while most still let AI agents share credentials—creating major cybersecurity risks as agent adoption accelerates. #AI#Cybersecurity#AIAgents#EnterpriseAI
📰 New York's ban on data center construction for a year has sparked concerns about its impact on the AI industry and may serve as a model for future anti-AI movements.
Evening! We’ve gotten lots of great feedback on the new ChatGPT desktop app (which we didn't get totally quite right on the first try), and as a result, we've made some changes.
The Cybersecurity Trap: Why 100 Tools Won’t Save You (But These 5 Skills Will)
Introduction: The modern cybersecurity landscape is often portrayed as a digital arms race, where the professional with the most extensive toolkit or the longest list of certifications wins. This perception, however,…
Huh, TIL: Firefox still provides critical security updates to Firefox 115 (version from mid 2023) for macOS as old as 10.12 (Sierra), until March 2027 😲 Very nice, @firefox.com! 👏
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Meet GPT-Red: an LLM super-hacker OpenAI built to make its…
Sources: Coatue is leading a $3B investment in Databricks that values the data analytics software company at $188B, a 40% increase from its December valuation (Wall Street Journal)
What TryHackMe Didn’t Tell You: The Darker Side of Your First CTF + Video
Introduction: Capture the Flag (CTF) competitions and platforms like TryHackMe are often viewed as the golden ticket into cybersecurity, offering a gamified path from theory to practice. However, relying solely on these…
Frontier labs want regulation, largely on their terms: The leaders of Google DeepMind, OpenAI, and Anthropic now agree on independent model testing and a national governing system, though they differ over enforcement.
Alphabet's Google will be required to open key Android features and share certain search-related data with OpenAI and other artificial intelligence rivals under new European Union rules aimed at curbing the company's market dominance, EU regulators said on Thursday.
From Zero to Network Hero: Mastering Network Scanning for Bulletproof Security + Video
Introduction: Before you can secure a system, you need to understand what is exposed. Network scanning is the foundational discipline that answers this critical question by discovering devices, services, and…
The latest update for #Acronis includes "How much does #HIPAA#compliance cost MSPs? A full breakdown" and "Managed IT services for health care: What MSPs need to deliver in 2026".
The latest update for #GitGuardian includes "The Perimeter Moved to the Laptop: From Network, to Identity, to the Developer Endpoint" and "Identity Infrastructure: Why Credentials Are the Layer Directories Don't Secure".
The latest update for #WatchGuard includes "Why Now Is the Time to Replace Your VPN" and "Why Human Behavior Has Become #Cybersecurity's Fastest-Changing Attack Surface".
📰 Apple is suing OpenAI for allegedly stealing hardware secrets through the recruitment of poached employees who brought over confidential presentations, secret prototypes, and key supplier details.
AlphaGEM enables precise genome-scale metabolic modelling by integrating protein structure alignment with deep-learning-based dark metabolism mining @natcomms.nature.com
Google stock dropped -4% today on reports that releasing Gemini 3.5 Pro has been delayed because it isn’t good at coding.
It’s notable that even Google’s stock price is dependent on whether its AI can replace software engineers. Must be fun to be in a career with such a huge target on its back.
Vamsee Pamisetty on Building Decision Intelligence That Stakeholders Can Actually Trust
Vamsee Pamisetty introduces an explainable AI framework that brings transparency, auditability, and trust to enterprise decision intelligence. #hackernews#news
What is Available: While waiting for the Pro flagship, developers and users can access Gemini 3.5 Flash. It is heavily optimized for speed and boasts a context window of up to one million tokens.
Get the most out of Google Workspace and streamline your daily tasks. Our newly updated Google Workspace with Gemini learning path gives you how-tos with more interactive lessons, a sense of humor and activities that take you from learning to doing with ease.
Source: Microsoft plans to release an AI security tool this month using models from Anthropic, OpenAI, and itself, as a cost-effective Mythos alternative (Aaron Holmes/The Information)
AI is more than a bubble, it’s the Hindenburg explosion all over again. Apple is putting Google’s Gemini and Siri both on every platform. They don’t need a black hole Data Center sucking down our resources.
Apple has reportedly contacted around 40 of its former employees who left the company for OpenAI, as it looks to gather evidence of alleged trade secrets theft by the AI juggernaut.
ASML plans to give its ~45K employees globally a €20,000 bonus, issued as a share award vesting in 2030, joining other chip industry companies offering payouts (Sarah Jacob/Bloomberg)
Information Security Briefly@infosecbriefly.bsky.social
Multiple cyber incidents across telecom, retail supply chains, manufacturing, and transportation caused data theft, operational shutdowns, and consumer notifications.
Ransomware claim says interlock hit Paragon Store Fixtures in the US, exposing internal design files, contracts, and client IP. The breach also revealed confidential project and architectural details. #UnitedStates#Manufacturing#Ransomware
“People are worried that if US companies start using Chinese models more and Anthropic less, then Anthropic will invest less. That means those US firms will lower the capex and in the end chip demand will be affected.” www.bloomberg.com/news/article...
CISA folds its own hard-won lessons into coordinated vulnerability disclosure guidance
On Wednesday, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) and four allied cyber authorities published a guide telling software vendors how to build a coordinated vulnerabil… #hackernews#news
"The [Linux] kernel is moving toward Rust. Git is moving toward Rust. Lots of projects are starting to move toward Rust": 'Rust makes coding fun again': Why Linux is moving away from C, according to Greg Kroah-Hartman buff.ly/YtyJjB3
Kimi K3 hype shouldn't alarm the US about "losing the AI race" to China as K3 is strong but not frontier-level and likely lacks dangerous cyber capabilities (Transformer)
OpenAI CFO Sarah Friar proposes a new AI scorecard with metrics for ROI, cost per task, dependability, and compute return. Useful for enterprise decision-making.
Attackers are using AI to move faster than ever. But you can stay ahead 🛡️
Join us at InterSECt 2026 to see how PAN-OS® 12.2 Ceres secures critical systems and neutralizes attacker infrastructure before an exploit even happens.
Asked #Google#AI a simple question yesterday, and the answer, as far as I can discern, was flat-out wrong: “The song "Fire in Cairo" by #TheCure is in the key of C♯ minor (C#m / D♭ minor).”
Information Security Briefly@infosecbriefly.bsky.social
UAT-11795 uses trojanized legitimate software and stealthy tools to steal credentials and cryptocurrency while maintaining persistent access across US and Europe.
This week may have changed everything
Apple sues OpenAI in possibly the worst case of corporate espionage ever, turns out Elon was stealing our code, and Kimi just earned it's Moonshot name!!!!
Here’s the actual Vilnius Town Hall displaying the Lithuanian and Ukrainian flags. An astute Bluesky follower caught the mistake generated by Gemini AI.
10 Best Profound Alternatives for Brand Visibility
Search is shifting from blue links to synthesized answers. In ChatGPT, Perplexity, Gemini and Google AI Overviews, your brand is either mentioned and cited or missing. I compared prompt tracking, citation monitoring, LLM coverage, benchmarking and…
"The model attempts to override the $HOME env var to define a temporary directory," said Sottiaux. "The model makes an honest mistake and mistakenly deletes $HOME instead."
CSIS Economic Security and Technology Department@est.csis.org
Is America ready for AI-powered cyberattacks?
CSIS Strategic Tech's Carol Kuntz and Lauryn Williams argue that the US needs to swiftly harden cyber defenses to protect against attacks launched by malicious actors using advanced AI models.
“All available indicators suggest that a ransom demand and ransom-related payment activity may have occurred,” the Parliamentary Joint Select Committee report on the 2023 cyber attack on the Government of Bermuda said.
The 6-Minute Supply Chain Heist: How a Compromised npm Package Turned Developer Machines into Credential Goldmines + Video
Introduction: On July 11, 2026, the software supply chain experienced a seismic shock when malicious versions of the jscrambler npm package—a JavaScript code-protection tool…
Apple’s lawsuit against OpenAI is likely to become one of the biggest litigations of our decade, but most of us will be more annoyed that Apple Music prices have gone up. All this and more in today’s Apple-related news summary. open.substack.com/pub/thecoren...
Track the evolving 2026 cyber threat landscape with our Major Data Breaches Timeline. We analyze initial attack vectors, data extortion tactics, and third-party supply chain failures across enterprise environments : cyberupdates365.com/data-breach-... #InfoSec#ThreatIntel#latest
999 SEC filings name Anthropic this year. It filed none of them. Its whole account of the offering ahead is two sentences, posted June 1. The piece reads what the filed record already holds against the page Anthropic will write. capefearadvisors.substack.com/p/anthropic-adding-it-up-the-judgment
Ending the week on The Vergecast, @reckless.bsky.social and I get into Apple v. OpenAI, the ups and downs of Siri AI, OnePlus bailing on the US, and Nilay's truly deranged emoji strategy
What the Oracle vulnerability says about today’s patching problem
The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has ordered federal agencies to secure their systems by Saturday against ongoing attacks exploiting a critical vulnerability in the Oracle E-Business Suite (EBS)…
‘Selling excess computing power to companies like Anthropic could provide Meta with a new revenue stream until demand for its own A.I. services catches up.’ Let’s hope!
GIFT LINK 1/3 www.nytimes.com/2026/07/17/t...
MCP, Agentic Mayhem, and the Quarter We Lost Control: Why Your Threat Model Is Already Obsolete + Video
Introduction: The security community has spent the first half of 2026 scrambling to secure the Model Context Protocol (MCP) even as agentic systems ship faster than anyone can inventory them.…
Join us and sponsor Veeam Software on July 29th for a FREE virtual summit on cloud resilience, data protection and incident response for lean IT teams.
Apple sued OpenAI for alleged trade-secret misconduct involving senior leadership and former Apple employees, potentially affecting OpenAI hardware plans and an IPO timeline.
