Daniel Miessler

article thumbnail

The AI Attack Surface Map v1.0

Daniel Miessler

Introduction Purpose Components Attacks Discussion Summary Introduction This resource is a first thrust at a framework for thinking about how to attack AI systems. At the time of writing, GPT-4 has only been out for a couple of months, and ChatGPT for only 6 months. So things are very early. There has been, of course, much content on attacking pre-ChatGPT AI systems, namely how to attack machine learning implementations.

article thumbnail

import alignment: A Library-based Approach to AI Alignment

Daniel Miessler

There is much being said right now about the AI Alignment Problem. There’s more to it, but it’s essentially the question of how to keep AI from destroying us when it wakes up. Yay. Russell argues that we need uncertainty built into our alignment efforts. I just read Stuart Russell’s excellent book on the topic—which I highly recommend—and Lex Fridman’s recent conversation with Eleizer Yudkowsky is an absolute must as well.

Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

article thumbnail

AI Revolutionizes Infosec

Daniel Miessler

After two and a half decades in information security, I’ve witnessed countless failures in security efforts. Most of them can be traced back to either asking the wrong questions or lacking the necessary context for effective action. Enter AI, which promises to address these shortcomings with two fundamental advantages: Context and Question-based security products.

InfoSec 168
article thumbnail

Calling Out The Security Community on AI

Daniel Miessler

I want to call out our community for a second on AI. And this applies to me as well because I have many of the same feelings. I feel there are too many in the security community who believe that AI is a minefield, and that it’s our job to warn people not to walk into it. I think our job is quite different. It’s not that people are considering walking into this minefield.

IoT 130
article thumbnail

Day-1 Skills That Cybersecurity Hiring Managers Are Looking For

Daniel Miessler

I’ve written before about the paradox of hiring managers not being able to find entry-level cybersecurity candidates, while many people with decent training or even degrees in the field cannot get hired. As it turns out, it’s not really that hard to explain. The Military takes you from zero to hero. In extremely large and long-term-focused organizations—like the Military—this gap is understood, so they spend time building extensive, standardized programs that can take any

article thumbnail

Unsupervised Learning: No. 192 (Member Edition)

Daniel Miessler

This is UL Member Content Subscribe Already a member? Login No related posts.

article thumbnail

A Conversation with Ken Modeste

Daniel Miessler

A while back I had the opportunity to chat with Ken Modeste, Director of Connected Technologies at UL, and I wanted to capture some of my takeaways from that conversation. I had the conversation as part of the media at Blackhat 2018. Here are some of the questions I asked: What is the mission of UL at this point? How does that relate to cyber? How are you different from other organizations with similar charters?

IoT 100