Adrian Sanabria

🎙️ Enterprise Weekly Podcast
🤝 Founder @bsidesknoxville
🗣️ Faculty @IANS_Security
🕵️ Security Research
🍳 Cooking
⛰️ Hiking
🏎️ F1

"I rant with data!"

2026-02-04

@g it tells me to scrape the pages manually and then give it the data, in both cases

I suspect AI browsers (headless or otherwise) must be configured to limit search/scraping depth to avoid DoSing websites?

2026-02-04

@g Like, the Claude Chrome extension, or OpenAI Operator?

Adrian Sanabria boosted:
2026-02-04

This stinks - @JosephMenn got caught up in the massacre of the Washington Post today.

A BlueSky message from Joseph Menn that reads: Most of the Washington Post’s tech reporters were laid off today, including me. I have loved my time at the paper, which is where I wanted to work from age 15. I take some consolation in not being among the survivors who will have to work harder with less for fewer readers. On to better things.
2026-02-04

Every now and then, I wonder: can AI do [insert simple task]? And I decide to let it have a whirl.

In this case, I wanted to make a list of IANS Faculty that were giving talks at this year's RSAC Conference.

The IANS faculty page is public: iansresearch.com/our-faculty/f

The RSAC Agenda page is public: path.rsaconference.com/flow/rs

Seems simple, right?

  1. Scrape site A
  2. Scrape site B
  3. Make a list of all people in list A that have a talk on list B.

It understood what it needed to do, but failed miserably.

  1. it cannot handle websites made for humans - for performance, each of these sites is paginated in different ways, to avoid loading all data all at once. It gets the first few pages and gives up after that.
  2. even with a small subset of data, it hallucinated and pulled in Hugh Thompson (he's not faculty).

Conclusion: I would be done with this task by now if I had ignored the existence of AI. Instead, I tried using AI, got tired and frustrated after trying half a dozen prompt approaches and different products (Claude, ChatGPT, and Gemini) and no longer have the time or energy to do it myself.

AI continues to be frustrating for these tasks that are too small to write a script for, but too large to want to do manually (open dev tools, copy JSON, manually extract/parse data on CLI, etc). Unfortunately, SO MANY tasks still require pulling information from a website built for humans, with no API.

2026-02-03

@mattblaze One of my favorite games of all time is Keep Talking and Nobody Explodes

keeptalkinggame.com/

It’s great - one person can “see” the bomb, the other has the “defusal manual”. The one with the bomb can’t see the manual, the one with the manual can’t see the bomb.

They have to work together and communicate well to get it done before the bomb. 5 minutes on the clock.

2026-02-02

@GossiTheDog many platforms have "innovated" by introducing ways of detecting, removing, and blocking AI-generated content. Pinterest, SoundCloud...

2026-02-02

@wendynather haha, I was in TAG in the late 80s, early 90s and oof, this was me aside from the oldest daughter bit

2026-01-31

@hal_pomeranz ooooh, I didn't realize this was a thing!

2026-01-30

@ferrix though, given his position, it’s a lot more like embezzlement than stealing

2026-01-30

@ferrix disgusting - literally stealing taxpayer money

2026-01-30
2026-01-30
2026-01-30

@chrismerkel that's elder abuse

2026-01-30

@dsalo agreed, good take, good context

the photo is especially on point

2026-01-30

@fwaggle @tehstu @hacks4pancakes we got some of those here (rebranded Holdens), right before the 2008 recession, for maybe 1 or 2 years before they were killed off because everyone is obsessed with giant trucks here

I think it was sold as the Chevy SS (marketing could have done better) and the Pontiac G8. No ute versions though, only 4-door sedans

2026-01-30

Talked about this on the podcast today

was way too proud of myself for coming up with this

2026-01-29

@Viss pretty sure this is the same law of physics that delivers the food to your table when you're in the restaurant bathroom and makes software bugs disappear the moment you connect the debugger

2026-01-29

@fwaggle @tehstu @hacks4pancakes let us dream that all Aussies are tearing around in utes powered by race fuel

2026-01-29

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst