Shellfyred

GFY. Slava Ukraini

@[email protected]@gfytwitter

[Shellfyred @Shellfyred]

@ScottAdamsSays @Jason Scott, how is it possible for you to talk with shit constantly streaming out of your mouth?

[Scott MacFarlane @MacFarlaneNews]

"I have never played a role in or rendered any services to a political party"

Todd Blanche submits 33-page questionnaire to Senate Judiciary Committee ahead of upcoming confirmation hearing

https://t.co/fZWGMJAZVv

[VS Code Changelog @VSCodeChangelog]

🚀 Insiders Update - Jun 11-Jun 13
9 new features

• ✨ Integrated Browser uses mapped host ports in Dev Containers
• Run Integrated Browser in SSH, tunnels, Dev Containers, and WSL
• Minimum release age option for extension installs and updates

🧵 See thread below 👇

🧵 1/3

[capitolhunters @capitolhunters]

IMHO it's a good general rule that if a club includes Jared Kushner, Peter Thiel, Elon Musk, and Leonard Leo, you should not join it. You do not want to be associated with those people in any way. And for those who did join, they'd better answer for that choice now. 7/

[Red Hat AI @RedHat_AI]

llama.cpp: runs on your laptop, no GPU required. GGUF files swap in seconds.

vLLM: built for concurrent users, GPU clusters, and production SLAs.

Same OpenAI-compatible API. Switching is a URL change.

New guide on picking the right one for your stack:
https://t.co/sgsVtFALJO

[Jamie Bonkiewicz @JamieBonkiewicz]

JD Vance promoting his book about finding his way back to faith while helping manage Situation Room meetings on how to cover up for pedophiles is about as Republican as it gets.

[Charlie Eriksen @CharlieEriksen]

I can no longer send security disclosures to a significant portion of the companies that are on my to-do list. Why?

Because they all use @Hacker0x01. My submissions so far get closed as informative, even though most fix the issue I report quite quickly. This makes my signal so low that I can no longer submit disclosures.

These companies have no other way to contact their security team. I'm now stuck with no way to responsibly disclose my findings to many companies. This is pretty messed up.

Not sure what to do from here. This sucks.

[Dan Bochman @DanBochman]

Typical coding day with Claude (Opus 4.8)

- explain to Claude the task (5 minutes)
- Claude implements task (10 minutes)

me: "Why is this necessary?"
Claude: "You're right to push back! I over-engineered this!"
- Repeat x87 times (13 hours)

[Luke Miani @LukeMiani]

> save $15M a year by cutting a screwworm monitoring program

> screwworm outbreak almost immediately

> $1B to combat it

Government efficiency

[respawnedtargaryen @respawnedtarga1]

It went from “porn” to everything like we all predicted

[Halvar Flake @halvarflake]

A short note on why I don't use the term "distillation" for training on the output of other models:

https://t.co/fs3ljeQ466

[Tim Miller @Timodc]

“Michelle Obama is a man” shouted on the White House lawn in a ring sponsored by Bud Light only available on Larry Ellison’s Paramount Plus. What a way to celebrate America 250 and the twilight of liberal democracy.

[RockWave @rockwavefm]

OLIVER TREE (1993-2026) 🕊️

[Roman Waschuk 🌻🍒🇨🇦🇺🇦 @WaschukCanUA]

As the Canadian official in charge of the bio labs file in 2007-8, I call BS. All of our activities, coordinated with the US, were above-board, and designed to PREVENT bioweapons proliferation, eg by replicating secure facilities like 🇨🇦 own Winnipeg lab in the post-Soviet space.

[Garry Kasparov @Kasparov63]

Loyal to the end. To Putin.

[patagucci perf papi @kenwheeler]

imagine if this technology fell into the wrong hands and foreigners could just one shot beige crud apps at will

[Shellfyred @Shellfyred]

What a dumb fucking post

[Halvar Flake @halvarflake]

Fable just downgraded to Opus because I am calculating a Groebner base on one round of the block cipher PRESENT.

This is absolutely ridiculous. I can essentially not use Fable to review my 2008 MSc thesis without triggering "cyber safeguards".

A friend of mine had the down...

[Florian Roth ⚡️ @cyb3rops]

NIST has a useful paper on AI guardrails

The takeaway is that static guardrails are the wrong security model for open-ended LLM systems.

A finite set of rules cannot cover every adaptive prompt. You can harden the system, make bypasses harder, monitor for abuse and reduce the blast radius. But you should not patch an LLM once, add a few refusal rules and call it done.

LLM security needs to look more like vuln research and detection engineering:
continuous testing, continuous updates and an assumption that bypasses will eventually be found

https://t.co/h2J7RIkPbn

[Nico Perrino @NicoPerrino]

Interesting history from Jane Kamensky in @TheFP.

Apparently Jefferson didn't think America should celebrate presidential birthdays.

"Vesting a national identity in individuals was what monarchies did, with profiles on coins and royal occasions organizing the state calendar."

[John Hammond @_JohnHammond]

happy @_ContinuumCon_ day to all who celebrate
https://t.co/RpEm2yB2rb