Patrick K.

AI is quickly becoming part of everyday developer workflows, and it’s changing the security model inside the IDE. Sharing a new post from Jeff Malnick, VP Engineering, Developer and AI at 1Password, on how AI agents expand the attack surface and what it takes to build secure, AI-assisted dev workflows. If you’re building or integrating AI tooling, this is a solid read: https://lnkd.in/gc3a9KEZ

12

Human error is inevitable, but a single misdirected email can cost your organization millions. Most tools focus on what’s in the message. Abnormal focuses on who’s sending it, and whether it makes sense for them to be sending it at all. With Misdirected Email Protection, Abnormal AI's behavioral AI understands each employee’s normal communication patterns to detect when an email is about to go to the wrong person—stopping accidental email exposure before it happens. Read our latest blog to learn how behavioral AI is redefining outbound email security.

8

Technology is political. I am noodling away at a project and wanted to understand something fundamental. Given that AI is here and technology will always be evolving and ever present: How do you feel about how you engage with it now and would you be interested in a tool that gives you more control over how you build context and how you protect the privacy and security of that context? In other words, do you care about how much personal and/or business information you share with various AI tools?

17

4 Comments

Question I've been asking clients lately: when's the last time you actually opened your WISP? Most firms wrote theirs once, filed it away, and haven't looked at it since. The problem is, a Written Information Security Plan isn't just paperwork. It's supposed to be the actual blueprint for how you protect client data and reduce cyberattack risk. A few reasons it's worth a second look: Regulators like the SEC and FTC actively audit for compliance. A thin or outdated WISP can mean real fines and legal exposure. A weak plan also means undefined protocols and no incident response plan, which leaves the door open to breaches. And honestly, a strong WISP builds client trust. It shows you take their data seriously. Rightworks put together a free WISP Analyzer that checks yours in about ten minutes. Swipe through below, and feel free to DM me if you want to talk it through. 👉 https://ow.ly/nNiJ50ZzOwx

14

This stat from our latest blog stuck with me. GitGuardian’s 2026 State of Secrets Sprawl report found 29 million secrets in public GitHub commits in 2025. Developers trying to work fast need convenience, but the easiest place to put a developer secret can sometimes be the riskiest place to leave it. Dropping a credential in an .env file in plaintext on disk is a risk that IT and security teams can't see, monitor, or revoke. That’s why I’m excited our team at 1Password is closing this gap. With enhancements to 1Password Environments and Developer Watchtower, we’re helping teams find plaintext .env on local disks and remediate exposed credentials in one click. Easily store, share, and use developer secrets across the tools and workflows developers already rely on. Security and velocity don't have to be a trade-off. Our latest blog dives deeper: https://bit.ly/4w0LcDn

2

AI can help patch software vulnerabilities, but it can’t reliably do the job on its own. Our Off-By-1 Labs team tested leading models on recent vulnerabilities they hadn’t seen before. Fewer than half of the fixes worked and some failed to resolve the issue, introduced new bugs, or made the problem worse. In a new episode of Zero Signal, 1Password CISO and CIO Jacob DePriest breaks down what the findings mean for teams using AI in security. Worth a watch: https://lnkd.in/gVczDKPG

1