Naresh LamGade retweeted
this is baller... nice one @censysio php2shell.censys.com/
Today was a big day for us š
Today, we onboarded our first beta customer to Vigile.AI Enterprise, an organization with 955 employee accounts.
@vigileai has already scanned 100,000+ files and analyzed sign-in and access logs, connected apps, user permissions, and sharing activity across their Microsoft 365 environment, uncovering previously unnoticed security risks and giving their team a clearer picture of whatās exposed, why it matters, and where to take action.
Vigile.AI connects to Microsoft 365 to help teams uncover sensitive data exposure, risky file sharing, and identity and access risks. It brings those findings together so security teams can understand what needs attention and where to act first.
Coming soon on Google Workspace.
Seeing something weāve worked so hard on running inside a real customerās environment feels special. Thereās plenty ahead of us, but today, we took a little time to celebrate together.
To our team: thank you for the hard work, patience, and dedication youāve poured into building this. And to our first beta customer: thank you for trusting us early and helping shape what comes next.
Cheers to the team, our first customer, and this beginning. š
Want to try Vigile AI Enterprise for your organization?
Request beta access at vigile.us
#VigileAI #Microsoft365 #Cybersecurity #StartupJourney
213
Great catch-up with my close friend @_jensec
We discussed open-source security research workflows, AI-assisted pentesting, multi-model research, sandbox testing, and how to approach complex targets from first principles.
The biggest takeaway:
Great security research is not about knowing more payloads. It is about understanding systems, challenging assumptions, forming better hypotheses, and testing them systematically and choosing the right model.
A lot of interesting work ahead around AI + cybersecurity + autonomous research workflows.
#CyberSecurity #BugBounty #AISecurity #SecurityResearch
1
212
Naresh LamGade retweeted
October is a reminder that cybersecurity is everyoneās responsibility.
This Cybersecurity Awareness Month, letās build safer digital habits and stay one step ahead of evolving threats.
Because cybersecurity starts with awareness.
š cynicaltechnology.com
Cynical turns 9 today! š
Feels like yesterday we started.
To our team, customers, and everyone whoās cheered us on: thank you for helping us come this far. ā¤ļø
Hereās to everything still ahead! š„
Today, we celebrate 9 years of innovation, resilience and growth. From humble beginnings to a trusted cybersecurity partner, this journey was built on trust, dedication and purpose. Hereās to the next chapter!
3
229
AI is helping people build faster, solve complex problems, and do more with smaller teams. Attackers are gaining those same advantages.
Anthropicās latest threat report shows how that is playing out from AI-assisted cyberattacks to concerns about where usersā sensitive data ends up. What stands out to me is how much a single person can now do with these tools. That changes the security challenges we all face.
Before adding an AI tool to your workflow, check where your data goes and who can access it. Keep passwords out of prompts, use approved tools for sensitive work, and grant only the permissions needed. Teams should also establish clear rules for AI use and reassess security whenever tools, integrations, or access change.
As we explore what AI makes possible, are we giving enough attention to the trust, privacy, and security that should come with it?
We're publishing our most detailed threat intelligence report to date.
It covers how people tried to misuse Claudeāfor cyberattacks, influence operations, surveillance, biology, and building weaponsāand how we found and stopped them.
We disrupted every operation in the report, and used the lessons from them to strengthen our safeguards. Where appropriate, we also shared what we found with authorities and other AI companies.
These cases are not typical: weāre highlighting some of the most sophisticated misuse weāve seen. But theyāre especially important to discuss, because they show us where AI misuse is headed, where our safeguards work, and where they need to improve.
Weāre publishing this report so others can spot the same activity on their own platforms, and so we can give the public a clearer view of how emerging threats develop.
Read the report: anthropic.com/threat-intelliā¦
140
Naresh LamGade retweeted
We've run a bug bounty program on Muse since early in development. Today it becomes public, with published payout guidelines. We pay by the impact demonstrated. More details at security.muse.ai.
Cyber is having a moment
Across 21 major software companies, including Apple, AWS, Microsoft, and Google:
- Reported critical vulnerabilities never cleared 100 per month in four years
- Since spring they've jumped to over 600 per month
Charts of the Week: a16z.news/p/chart-of-the-weeā¦
Naresh LamGade retweeted
Inviting Applications from Indian Social Media Influencers/Vloggers/Podcasters for Familiarization Trip to Nepal
Naresh LamGade retweeted
Introducing Project Glasswing: an urgent initiative to help secure the worldās most critical software.
Itās powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans.
anthropic.com/glasswing
RSA 2026, here we come!
RSAC 2026
Where the world's best security minds meet, Vigile AI will be there.
RSAC 2026 is the perfect stage to show the world what AI-driven security truly looks like.
See you in San Francisco.
#RSAC2026 #RSAConference #RSAConference2026 #VigileAI #DarkWebMonitoring
2
245
Naresh LamGade retweeted
Google API keys didn't use to be considered "secret," so they're all over the web-- but now they are an open door to Gemini š« Quick rundown video of Truffle Security's really nifty research, almost 3,000 websites exposed.. including Google themselvesš
š youtu.be/XNMHUifKce8
Naresh LamGade retweeted
Can AI beat challenges modeled after actual prevented breaches found in the wild? They did - 9 out of 10 times - and usually for less than a dollar.
The new economics of cyber offensive AI are hard to ignore.






























