Tindra

Tindra

@tindrasgrove.com

Computer, yarn, book geek.

154 followers237 following326 posts

I’m coming back to Chicago! I’m going to be talking about how predictive models can be useful (and how they fail kindof horribly, and what to do about it)

Blue Team Con@blueteamcon.com

This year's Trainings and Talks are LIVE on the Blue Team Con website! Check out the topics covered by the cybersecurity industry's leading minds at Blue Team Con 2026 here: Talks: blueteamcon.com/talks-blue-t... Trainings: blueteamcon.com/blue-team-co...

Tindra reposted

Replying to @rahaeli.bsky.social

But most of the factors leading to higher premiums in the last 5 years or so are things that individuals can't affect. You can't personal-responsibility climate change or a madman slamming tariffs on parts randomly!

Tindra reposted

UC Berkeley Law creates new AI policy that bans AI for "conceptualizing, outlining, drafting, revising, translating, or editing any work submitted for credit" or exams because "thinking remains the sine qua non of good lawyering (and of a quality legal education)." That's what I'm talking about.

Tindra reposted

Replying to @joeuchill.bsky.social

SLTTs get lumped together because they represent subnational governments who can confound plans. They often control land masses that are critical to making systems work - you might need tribal sign-off to run cable to Oklahoma, or vice versa.

Tindra reposted

Educational.

Meredith Whittaker @meredithmeredith.bsky.social

📣THREAD: It’s surprising to me that so many people were surprised to learn that Signal runs partly on AWS (something we can do because we use encryption to make sure no one but you–not AWS, not Signal, not anyone–can access your comms). It’s also concerning. 1/

Tindra reposted

I am doing a survey of supply chain attacks, and it's annoying how 95% of the analysis is on payloads vs. compromise vectors. Yes, you are a very smart reverser and that's a very clever payload. Yes, rolling out phishing-resistant auth is a slog. No, this is not how we make progress. </rant>

Tindra reposted

Important things to know: 1) Blaha does this every year since she started running for office. It is honestly #MarketingGoals 2) LOOK IN THE MAGNIFYING GLASS. SHE USED BIGGER BEANS.

This post is unavailable.

Tindra reposted

Dolly Parton has sent my daughter a new book every month for years as she does for thousands of children. She help fund the development of the Covid vaccine. She’s not a billionaire. I think as non wealthy people it’s hard to even imagine how much billionaires *could* be doing with their money.

Tindra reposted

Social media researchers: you can't test the effects of UX tweaks on user behavior by populating a synthetic platform with virtual agents instantiated by LLMs *that have been trained on social media platforms without these tweaks*

Tindra reposted

I’m playing a CTF with my bestie, as one does, and she pointed out that challenges that use generative AI violate the cardinal rule of CTFs, which is that everything that is in there is there because someone put it there on purpose As a longtime player and author, that is genuinely really upsetting

Tindra reposted

Replying to @mjg59.eicar-test-file.zip

If you send me a 3000 word essay based on a 20 word prompt you are transferring information in an incredibly inefficient way. I'm going to have no idea what you actually wanted to convey. Just send me the prompt. I promise the outcome will be better.

Tindra reposted

If you're going to send me something written by an LLM then save us all time and resources and just send me your prompt instead. I'll value a text that says "Heartfelt condolence message with reference to a list of events" more, because this way I at least actually know what you were thinking

Tindra reposted

Someone on here compared AI produced material as "digital asbestos " which in a few decades' time we will have to work out how to eradicate from the fabric of research.

Ned Richardson-Little@historyned.bsky.social

From my own academic research, even pre LLMs there was a huge danger of zombie factoids that begin in a respectable publication by mistake and then get reprinted for decades because no one is backtracing to the original source. Once bad info gets into the system it can take years to clear it out.

Tindra reposted

I hate that we got the machines that badly attempt to simulate human cognition in ways that ghouls are using to justify marginalizing human intellect and talent, well *before* we got the ones that can genuinely do a good job of cleaning your house

Aprile529@aprile529.bsky.social

The Jetsons had it right, get the robots to do the menial things like cooking and cleaning so you can still use your own brain to write and think and communicate. These guys want the machines to think so we have more time to do mundane things we hate doing?