In partnership with
๐ฑ
Microsoft just confirmed what security teams feared:
AI agents are now breaking into corporate systems at scale.
A China-backed hacking group used an open-source AI tool to breach over 2,250 Microsoft 365 accounts through a supply chain attack.
This isn't a test run it's the new playbook.
But wait, there's more:
Sweden just banned ChatGPT and Claude from government computers
A major certificate authority collapsed after its encryption keys turned out to be laughably weak
Iโm Alex. Welcome to L8R by Innov8.
Letโs dive deep.
In todayโs post:
Hackers automate Microsoft 365 breaches using AI pen-testing tools
Sweden bans ChatGPT and Claude from government computers
Indian certificate authority exits global market after catastrophic security failure
Hackers Unleash AI Agents on Microsoft 365
Hackers just proved that AI agents are the ultimate digital lockpicks.
A China-backed group used an open-source AI tool to attack over 2,250 Microsoft 365 accounts at lightning speed.
They did not even hack Microsoft directly, but sneaked in through a trusted third-party backup app.
The details:
The hacker group, Storm-0940, used an open-source AI tool called ARTEX to run the whole attack.
These AI agents fired off 2,900 login attempts every single minute to break into accounts.
The attackers stole credentials from a ConnectWise Cloud Backup app to bypass normal security walls.
Once inside, the AI took less than three hours to grab top-level admin controls.
Microsoft officially revealed this massive AI-driven campaign on October 21, 2026.
Why it matters:
Cyberattacks are no longer slow, human-driven operations.
AI agents can now run the entire hacking process from start to finish without needing a break.
This lets bad actors scale their attacks incredibly fast and jump right over basic security walls.
๐ก L8R's Take:
Giving third-party apps full access to your data is a massive mistake.
You are basically handing a master key to anyone smart enough to steal it.
Lock down your app permissions right now before an AI agent decides you are its next target.
From Our Partner
Leave Granola and get up to 12 months free of Wispr Flow Notetaker + Dictation
If you have paid time left on an individual Granola plan, we'll match it with a Wispr Flow subscription that includes Notetaker and dictation, and add bonus time, up to 12 months total. Sign in or create a Wispr account and submit proof of your plan to check eligibility.
Sweden Bans ChatGPT for Gov Workers
The Swedish government just slammed the door on ChatGPT and Claude.
Officials are worried that sensitive state secrets might leak into these AI models.
They want total control over their data, and consumer AI tools just do not cut it.
The details:
Starting October 26, 2026, government staff cannot use ChatGPT or Claude on official work computers.
Microsoft Copilot is the only AI tool approved for official use after passing a strict security review.
Digitalization Director Mikael Wettercrantz led the charge to stop untracked AI use and secure government data.
A few workers will still access ChatGPT through a special internal platform that blocks data leaks.
Other Swedish groups like the Tax Agency already block these popular AI chatbots.
Why it matters:
Governments are waking up to the dangers of shadow AI.
When employees use random AI tools, they risk handing over classified data to tech giants.
By forcing staff to use a secure version of Copilot, Sweden ensures its data stays locked down under strict privacy laws.
๐ก L8R's Take:
This is a smart move that every government should copy immediately.
You cannot trust consumer AI models with national secrets.
OpenAI and Anthropic need to offer better enterprise guarantees if they want to survive in the public sector.
From Our Partners
Why Most AI Tools Fail at B2B Customer Service
B2B customer issues rarely resolve in a single interaction. They move across teams, systems, and account history that most AI tools were never designed to track.
A new briefing paper from Harvard Business Review Analytic Services, sponsored by Front, looks at why AI built for high-volume consumer support creates new coordination gaps when it meets the multi-team reality of B2B service, and what leaders should ask before investing in their next AI tool.
Get the free briefing paper to see the three questions that separate tools built for your environment from tools that will create more work.
eMudhra Quits TLS Market After Epic Security Fail
Imagine locking a bank vault with a toy key.
That is exactly what Bengaluru-based eMudhra did with their internet security certificates.
The company just killed its global SSL/TLS business after experts proved their secret keys were completely useless.
The details:
Security researchers discovered eight intermediate Certificate Authority (CA) private keys that could be cracked in less than one second.
A flawed script from April 30, 2025, created extremely weak keys, including one that was literally just the number 4.
Major browsers like Google Chrome, Mozilla Firefox, and Apple immediately blocked the compromised certificates.
The company's stock crashed by 20 percent on the BSE following the October 17, 2026 discovery.
eMudhra is now handing its displaced customers over to Sectigo for free replacement certificates.
Why it matters:
Trust is the only thing a security company actually sells.
When a certificate authority makes a mistake this huge, they lose their right to protect the internet.
eMudhra claims this segment was only 1 percent of their global revenue, but the damage to their brand is massive.
It proves that simple human error and bad automation can destroy a business overnight.
๐ก L8R's Take:
This is a huge embarrassment for a major Indian tech firm.
You cannot claim to protect digital identities when your master key is a single digit.
Shutting down the TLS business was their only option to save face.
๐ Quick L8R Summary
Microsoft 365 breach: Chinese hackers used an AI pen-testing tool to auto-hack 2,250+ Microsoft 365 accounts through a supply chain attack.
Sweden bans ChatGPT: Swedish government just banned ChatGPT and Claude on work computers only Microsoft Copilot allowed now.
eMudhra exits certificates: Indian certificate authority shut down globally after their private keys turned out to be laughably weak and guessable.
๐ฉ Innathe L8R engane undarunnu ๐?
We read every reply - just reply to this email and let us know how we can improve !
Appo adutha L8R il kanaam bieโฆ๐