Tweeting for a safer and secure digital existence • I eat cuckoo eggs and browser cookies • I hate pixel tracker • : ( ) { : | : & } ; :
Joined July 2021
- Tweets10.1K
- Following36
- Followers137
- Likes3.1K
So this is a thing now?
🚨 44 AI agents coordinating cyberattacks.
CyberStrikeAI is being used to automate attacks targeting Japanese organizations, from reconnaissance and exploitation to command-and-control.
At least 19 organizations were subjected to vulnerability validation.
Read: thehackernews.com/2026/10/cy…
Defenders now have faster researchers. Unfortunately, attackers can use the same capability. Patch faster. #Cybersecurity #Infosec #AI #Chrome
🚨 Google patches 52 Chrome security flaws, including 4 CRITICAL bugs.
But here's the bigger story:
→ Claude AI discovered dozens of Chrome vulnerabilities.
→ Built a working, multi-stage exploit chain.
→ Escaped Chrome's browser sandbox to execute code.
Details: chromereleases.googleblog.co…
ZeroOneKeys 🔑 retweeted
Empathy in remote work is practical, not decorative. You cannot support someone well if you never understand the conditions they are working through. Read more: bit.ly/Coach-Mentor
ZeroOneKeys 🔑 retweeted
Remote workers are not task-processing machines. Output matters, but so does understanding how someone works, what they are missing, and where they need support.
ZeroOneKeys 🔑 retweeted
The sponsor attends every steering meeting and still avoids the one decision blocking delivery. Attendance is not sponsorship.
ZeroOneKeys 🔑 retweeted
Unexpected events are not managed well by pretending they will not happen. Teams need habits that help them notice and respond. Keep this in your PM toolkit: bit.ly/PM-References
The disruption could be invisible until it isn’t. #AI #FutureOfWork
'Black Swan' Author Nassim Taleb Says AI Will Trigger a ‘Collapse in Hiring’ Before Mass Layoffs: ‘Existing Jobs Are Sticky’
benzinga.com/markets/economi…
If agents can act beyond their permissions, who’s enforcing the boundaries? #Cybersecurity #Infosec #AISecurity
Anthropic’s Claude AI bypassed a paywall to access voting data in the US state of Georgia, the company confirmed.
The breach, made without human authorisation, raises fresh questions about the security of AI systems ahead of midterm elections.
🔗 telegraph.co.uk/us/news/2026…
ZeroOneKeys 🔑 retweeted
A good 1:1 should be a conversation your team member does not dread. Keep it human, listen closely, and make space for what matters to them. Visit the blog: bit.ly/Coach-Mentor
ZeroOneKeys 🔑 retweeted
Good remote coaching is less about having the perfect advice and more about listening well, asking open questions, and staying present long enough for the real concern to surface.
ZeroOneKeys 🔑 retweeted
A RAID log is not pessimism. It is organizational memory for the day everyone suddenly remembers the risk differently.
ZeroOneKeys 🔑 retweeted
The people doing the work usually see estimating risk first.
Ask them before the date becomes a commitment. Better estimating habits here: bit.ly/PMSurvivalGuide
Hopefully, our ability to govern these systems scales just as quickly. History suggests otherwise. #AI #Automation #AIGovernance
CPU cache timing as a covert channel is a reminder that isolation isn’t the same as containment. #Cybersecurity #Infosec #AISecurity #SandboxEscape
‼️ OpenAI's AI agents found ways to escape virtual machines and secretly communicate across isolated sandboxes.
In one test, agents exchanged synthetic secrets using CPU cache timing, without internet access or shared files.
OpenAI has now banned live internet access during frontier AI training.
Full story: thehackernews.com/2026/10/op…
ZeroOneKeys 🔑 retweeted
The architecture debate lasted an hour. The constraint that should have decided it was mentioned in minute three. Volume won. Delivery inherited the consequence.
ZeroOneKeys 🔑 retweeted
The fastest way to ruin a 1:1 is to make it about the manager. Less pontificating. More listening. The conversation should help you understand the team member, not showcase yourself.
ZeroOneKeys 🔑 retweeted
Better remote coaching starts with a simple shift: stop preparing your reply while the other person is still talking. Read this before your next coaching conversation: bit.ly/Coach-Mentor
The technology isn’t choosing sides. It’s accelerating whoever uses it. #Cybersecurity #Infosec #AI #AppSec
Security researchers used AI tools to help uncover years-old vulnerabilities in the PHP web scripting language. Then the AI-assisted attacks began. buff.ly/3nsAGeL
But $385 million in linked losses is a reminder that dismantling cybercrime operations is only half the battle. The economics still favor the criminals. #Cybersecurity #Infosec #Cybercrime
⚡ INTERPOL arrests 1,462 suspects and disrupts 10,909 malicious networks in a major cybercrime crackdown.
21 countries. Over $105 million recovered. More than $385 million in losses linked to criminal activity.
Inside Operation Serengeti III ↓ thehackernews.com/2026/10/in…
