Lojiholia enshrines all triumph's Avatar

Lojiholia enshrines all triumph

@lojikil.bsky.social

Brains in the "trying to be a good dad despite having a bad dad" gang. ☦️|Father|Philosopher|Offensive Security|PLT 
I work in security at a decent sized firm, but my research interests are actually around Multics, ML dialects, &c

135 Followers  |  307 Following  |  46 Posts  |  Joined: 08.05.2023  |  2.4895

Latest posts by lojikil.bsky.social on Bluesky

Preview
Enough of the billionaires and their big tech. ‘Frugal tech’ will build us all a better world | Eleanor Drage Titans like Musk would love us to believe innovation means top-down solutions that only enrich the wealthy. In fact, we all have the power, says Eleanor Drage, research fellow at Cambridge University

Excellent. www.theguardian.com/commentisfre...

21.09.2025 10:53 — 👍 7    🔁 4    💬 0    📌 0

Hello hello! I’m worse with responding on socials nowadays, but I’ve always enjoyed our conversations! How are you doing??

30.05.2025 19:59 — 👍 2    🔁 0    💬 1    📌 0

Ja I think this is it; I’ve largely stopped posting anywhere, but if I do it’s mainly Mastodon (for whatever that is worth).

27.05.2025 00:12 — 👍 4    🔁 0    💬 0    📌 0

And you as well, really great to see what you’re working on now!

Also, I’m sorry for the delay, I don’t really use any social media anymore.

27.04.2025 23:08 — 👍 0    🔁 0    💬 0    📌 0

I really love when a language or tool you’re working on comes together like this. Great work!

25.04.2025 00:15 — 👍 0    🔁 0    💬 0    📌 0

We need a @badthingsdaily.bsky.social entry for “your most senior advisors have invited a journalist to a signal chat they’re not supposed to have”

25.03.2025 14:50 — 👍 2    🔁 0    💬 1    📌 0

I’m pretty sure this is us everyday, friend

22.03.2025 14:29 — 👍 2    🔁 0    💬 0    📌 0

Congratulations! Are you focusing on any specific part of Startups or a specific vertical?

18.03.2025 23:33 — 👍 0    🔁 0    💬 1    📌 0

We asked this machine trained on war scenarios for how to end the conflict in Ukraine, and you’ll be shocked to learn how it turns out!

04.03.2025 19:13 — 👍 1    🔁 0    💬 0    📌 0

Also didn’t we miss the lessons of Toyota’s lean delivery? Iirc they had much more planning, supply chain depth, and understanding of capacity and utilization than most “JIT and JIT alone” proponents expound

22.02.2025 20:56 — 👍 1    🔁 0    💬 1    📌 0
Preview
Cutting the Nuclear Weapons Enterprise Serves No Strategy - The Council on Strategic Risks Weakening the enterprise benefits neither a build-up or build-down approach and potentially ties the hands of negotiators.

LATE from me: You know you've done some weird stuff when *everyone* in the nuclear weapons field is upset.

There are a few paths forward with a diminished DOE + NNSA, but none of them properly support an build-up, the status quo, *or* reductions. Another massive own-goal no matter your priors.

14.02.2025 23:23 — 👍 95    🔁 29    💬 2    📌 7

Two comments:

First, item (a)(iii) would be financially ruinous for us—even if it can be made to work—not least because we would constantly have to maintain vast numbers of space-based interceptors in orbit to ensure enough are always on station high above Russia, China, North Korea, all ...

28.01.2025 16:36 — 👍 17    🔁 8    💬 4    📌 1

And I still maintain that this is big “mineshaft gap” energy. The handwringing about how “China might be fielding LLMs on the battlefield” is just wild to me. If you want to talk information superiority/dominance, maybe, but otherwise it seems like GEN Turgidson pointing at the big board

28.01.2025 15:02 — 👍 0    🔁 0    💬 0    📌 0

I hope @edzitron.com writes more about this, but the assumption that lighting the earth on fire for some sort of NOBUS AI capability that China/Russia can’t have is tied to not understanding the tech or not understanding great powers competition or both (likely with an aperitif of jingoism)

28.01.2025 14:58 — 👍 4    🔁 0    💬 1    📌 0

I wish the US did this, especially if it were at the Caucus level

26.01.2025 19:52 — 👍 0    🔁 0    💬 0    📌 0
Preview
Schools Using AI Emulation of Anne Frank That Urges Kids Not to Blame Anyone for Holocaust A Utah-based startup has graced us with an AI-powered Anne Frank chatbot, and it's very eager to forgive and forget.

Part of the overall project with these systems in addition to eliminating critical thinking, is to water down anything transformative or radical.

19.01.2025 01:17 — 👍 2293    🔁 751    💬 113    📌 405
Preview
Data Prior to Language If the author is dead, why isn't the LLM?

Some notes on Barthes’ “The Death of the Author,” the false authority of LLMs, and why you can’t argue that the author is dead if you’re using AI to reanimate the corpse. mail.cyberneticforests.com/data-prior-t...

12.01.2025 13:07 — 👍 49    🔁 11    💬 9    📌 8

Arguing with a toddler as to who is delirious at 0540 when he has been up and down for two hours is fun.

Me: go to sleep bud you’re delirious
Him: nO yOu ArE dElIrIoUs

11.01.2025 10:43 — 👍 0    🔁 0    💬 0    📌 0

It’s pretty amazing to me that any sort of remote management is allowed that could reach into these sorts of systems. Not surprising per se, but still amazing

06.01.2025 20:49 — 👍 1    🔁 0    💬 0    📌 0

I definitely didn’t expect Santa vs Ded Moroz dueling propaganda wars…

30.12.2024 22:38 — 👍 1    🔁 0    💬 0    📌 0

Now the party is ready to get started…

30.12.2024 22:27 — 👍 1    🔁 0    💬 0    📌 0

A game I had installed on my iPad just asked if I would agree to arbitration; in light of Disney, I’d be curious how far and wide such agreements will eventually cover.

28.12.2024 16:55 — 👍 0    🔁 0    💬 0    📌 0

A brutal regime falls. But there’s a problem: the UN and several countries say the victorious rebels are terrorists. What to do? There’s no playbook for Islamist insurgents faced with the challenges of running a country. But there is a precedent, in Afghanistan. 1/

09.12.2024 12:02 — 👍 76    🔁 43    💬 6    📌 15

Ja I spent a lot of time reading these sorts of scams at my last job, it was fascinating how quickly actors could adopt to a new security control

03.12.2024 23:29 — 👍 0    🔁 0    💬 0    📌 0

It’s also interesting how they updated for filtered messages, suggesting you should copy and paste the link if it’s not clickable (which it won’t be for the Unknown Senders filter)

03.12.2024 22:51 — 👍 0    🔁 0    💬 1    📌 0
Preview
Exclusive: Exxon lobbyist investigated over hack-and-leak of environmentalist emails, sources say The FBI has been investigating a longtime Exxon Mobil consultant over the contractor's alleged role in a hack-and-leak operation that targeted hundreds of the oil company’s biggest critics, according to three people familiar with the matter.

Just before the holiday, we broke the story this weekend of how a longtime Exxon lobbyist has been investigated by the FBI over a sweeping cyberespionage campaign targeting hundreds of environmentalists across the United States.

www.reuters.com/business/ene...

02.12.2024 13:47 — 👍 37    🔁 15    💬 1    📌 3

Also, of further issue, it looks like none of the records changed in the ETL for the v4 or v5 CVEs.

So basically, it looks like something wonky happened with a bunch of early records during a migration or the like internally? /cc @jgamblin.bsky.social

02.12.2024 21:42 — 👍 0    🔁 0    💬 0    📌 0

So spot checking a few and it looks like nothing of substance changed…

02.12.2024 20:08 — 👍 0    🔁 0    💬 1    📌 0

I guess I’ll audit and see if they *actually* changed or this was just a reload. This sort of data hoarding is why I have like 80GB of vulnerability data

02.12.2024 18:22 — 👍 0    🔁 0    💬 1    📌 0

… but I think they may have screwed up the data? Requesting everything updated since 2024-NOV-20 returns a ton of 1999 data…

02.12.2024 17:19 — 👍 0    🔁 0    💬 1    📌 0

@lojikil is following 20 prominent accounts