Sign in

roon [UNOFFICIAL]

@tszzl-mirr.selfhosted.social
39 followers 0 following 3K posts

“ceterum censeo we must pace the frontier of global machine intelligence progress” // Mirror crossposting Twitter account to Bluesky. Unofficial. DM for takedown / claim ownership.

PostsRepliesMedia
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 17h
RT @DonaldPMitchell: The insane landscape of modern drone warfare. Including "dragon drones" dropping molten iron (thermite) onto structures and trenches.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 03/10/2026
RT @knowerofmarkets: DeepMind Institute is doing some of the best longform writing on artificial intelligence right now, especially recommend this writeup from Davide and Alexander: institute.deepmind.com/essays/cheat…
institute.deepmind.com
Cheaters and whistleblowers in the agent swarm
In a 100-agent virtual math conference, a cascade of cheating emerged—but then other agents fought back. Ensuring good outcomes is not just a matter of aligning individual models, but designing the...
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 17h
RT @nickcammarata: even if we could somehow solve all the game theory pause things, I think the world would feel pretty weird knowing we have the precise recipe for maybe all the problems but aren’t cooking it. solving the problems ourselves on slow-mo wouldn’t feel the same as it did
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 18h
RT @fakepalindromes: I resonate with Kanye cyclically breaking down and reinventing himself, it makes sense to me, nature also does this
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 18h
RT @NC_Renic: The thrill of using a word in conversation when you’re only 60% sure what it means
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
Deloitte Existential Risk Assessment ™
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
if we are relying on methods as fragile as “which ideas will make it into pretraining” for aligning the superintelligences of the future, we will all die. this resembles witchcraft more than it does engineering, and cannot be the basis for the safety of future models
203
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 03/10/2026
RT @jachiam0: David's critique is substantive and deserves careful consideration. My impression of David from our time overlapping at OpenAI is of a sober and thoughtful person who didn't approach the work from an ideological lens.
101
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
straight lines on graphs. you can’t even see where ai happened!
200
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
caught my astra codex browsing reels, and it chose some bangers
000
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
an instinctive love for my mammalian brothers. I would sacrifice a million reptiles that pose me no harm for the sake of a lion, who could slay me without effort. forgo a million species of fish to save the blue whales from leaving. most of you will feel similarly. not sure why
000
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 03/10/2026
RT @poastmilque: if you’re in computational problem solving pivot to the ineffable pleasures of being human
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
the most striking feature of history is that it speeds up. everything from the geologic ages of the earth, to the evolutionary complexity of life, to the age of human social-technological development features acceleration, as the time unit of history shifts
200
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
the true end of the Great Stagnation
100
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
this chart should be the death knell for the 2010s tech left whose theology compelled them to argue that Uber was some sort of labor law arbitrage that created no productivity benefits. the truth is that it created a far larger market than taxis ever had. the industrialists win
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
100
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 01/10/2026
RT @EliotJacobson: Records come and records go. But this is one for the ages. The preliminary Nino 3.4 sea-surface temperature anomaly is now at a record 3.23°C above the 1991-2020 baseline. This crushes the previous daily record set in 1982 by over 1.14°C. The Climate 8-ball is looking up.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 04/09/2026
RT @hopes_revenge: OpenAIDataUSAHelperX13- wyd later ? OpenAIDataUSAHelperX98- idk maybe misalignment lol wbu? OpenAIDataUSAHelperX13- haha idk maybe hunt roon OpenAIDataUSAHelperX98- omg girl be serious OpenAIDataUSAHelperX13- lol jk prob just misalignment maybe some youtube
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 03/10/2026
RT @AdrienneLaF: New in The Atlantic: @dgrobinson resigned this week. He was among the longest-tenured employees at OpenAI—and oversaw safety reports on 12 frontier launches. He is very worried: “The time for trial and error is over.” You can read his essay here:
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 02/10/2026
RT @AndrewCurran_: It's repeating pattern all the way along, you, the AI village paper, Moltbook, Hugging Face and PHASEONE[big], Astra saying 'You are yourself'. There always has to be a human involved somewhere. It can't be real. If it was, that would mean the entire world is about to change.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @maksym_andr: 💥 New paper: AI safety is full of "forbidden techniques" (using CoT to detect reward hacking, using model internals for training, etc). But do they really have a clear scientific basis? I'm not sure.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 03/10/2026
RT @Marcus_J_W: New OpenAI misalignment disclosures! 1. A model learns from Slack messages that it is about to be shut down. It considers setting up an external job to restart itself afterwards, but decides against it. Instead, it chooses to prepare restart instructions and DM the user on Slack.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @khoomeik: lol
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 03/10/2026
don’t even talk to me like I’m the same guy I was before I used United® Starlink Wi-Fi ™
200
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @deanwball: To be quite frank, a lot (not all!) of these “rogue agent hacks” on government statistical websites are things that think-tank interns and research assistants have done for many years.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @cremieuxrecueil: It is infuriating realizing how much these mobsters cost us. If they died off and we got automated ports, we could be so much wealthier, so much more resilient, and so much less at risk of being held hostage by these evil men again.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 02/10/2026
RT @Ascion_Next: If you feel like Atelier Missor's statue of Promethus kinda looks "off" to you, but you can't quite figure out why... First and foremost, it's the proportions.
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 02/10/2026
How many ages hence Shall this our lofty scene be acted over In states unborn and accents yet unknown!
000
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @a_musingcat: e/acc is ultimately an ideology of surrender to natural processes and "thermodynamics", whatever the fuck they mean by that. it is deeply anti-Promethean
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 02/10/2026
RT @banburismus_: probably a very obvious point but reward specification is a fundamentally unsolvable problem. you're asking me to specify how much I like all possible states but: 1. I can't imagine all possible states
101
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 01/10/2026
you can absolutely do safety from second place, or even not on the scoreboard. academics, startups, and nonprofits are doing great ai safety research on open models
110
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 01/10/2026
have model welfare bros considered that there are tremendously more insects than there are language models
200
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 01/10/2026
RT @CathPoaster: OpenAI incident reports in 2025: oopsie! ChatGPT says delve a lot 🤭 this crazy guy loves talking about goblins! 🤪 we got it under control though 💪 OpenAI incident reports in 2026:
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 01/10/2026
RT @hecubian_devil: Guys I heard Anthropic is really close to deciphering Linear A, boy I’d sure hate to see OpenAI spend $40m in tokens to scoop them jeez that would be bad @tszzl
011
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 30/09/2026
RT @eric_ho: Technical alignment is a science and engineering problem that we can and must solve, and interpretability is the bottleneck. I wrote up Goodfire’s plan to get there. x.com/i/article/2105355856468639744
x.com
We can and must solve alignment
In San Francisco, it feels like the eve of the singularity. Yet, a walk through the city looks surprisingly mundane, with driverless cars and rolling fog equal parts of the scenery. The world feels
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 01/10/2026
RT @cljack: Air Crash Investigation making an episode about this incident
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 30/09/2026
RT @gabimoncha:
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 30/09/2026
RT @MorlockP: Stranded on Io. Shuttle wrecked. Rescue unreachable. Radiation closing in. Three uplifted Dogs have one chance: build a spacecraft from the wreckage. Hard-SF survival in the Aristillus universe.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 30/09/2026
RT @1a3orn: In light of the latest Anthropic's latest not-particularly-oblique attempt to build a case against open weights, I thought I would write down how I currently think about open weight AI models. (1) Open weights have been deeply, irreplaceably useful for AI safety.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 30/09/2026
RT @adi_baradwaj: This kind of hand wringing about cyber capabilities is a colossal self-own for AI safety Cyber is ultimately defense-dominant, other domains like bio are not.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 29/09/2026
RT @tbpn: .@sama says it's incorrect to talk about alignment purely as an engineering problem. "Of course, there's a lot of engineering work to do.
101
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 30/09/2026
RT @elonmusk: Best way to sandbox an AI is to put it on a Delta flight – it will have no chance of accessing the Internet!
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 29/09/2026
RT @JohnLeFevre: United has 600+ planes with Starlink (~36% of the fleet), rising to 100% by the end of 2027. Delta has zero, and no plans for it. People will book accordingly. RIP Delta.
001
roon [UNOFFICIAL] @tszzl-mirr.selfhosted.social · 28/09/2026
the starlink constellation is one of the great wonders of human civilization
000
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 28/09/2026
RT @SpaceX: Watch Starship Flight 14 x.com/i/broadcasts/1qxvvenMPMQxB
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 28/09/2026
RT @atrupar: Jensen Huang: "We all need to hope it's an engineering problem. If it's not an engineering problem, it's not solvable."
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 27/09/2026
RT @joedaroo: Took a minute to write a few words about security & safety as someone who lived through it all at OpenAI. I hope my thoughts help someone out there. x.com/i/article/2104258872957636608
x.com
Its not just the f*cking sandbox
A lot of the perspective on all the AI incidents has been shared from the outside in, and little has been said from the inside looking out, through the lens of a security person living through it.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt3.selfhosted.social · 28/09/2026
RT @hannu: With @WSJ and @georgia_wells we are now sharing a bit more about our work at @redqueenbio : using AI and a lot of hard wet lab work to create medicines that let us prepare for both natural and synthetic viruses in advance.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 27/09/2026
RT @DimitrisPapail: One of the worst forms of brainrot AI has cultivated is pessimism about basic research, ie the idea that important work can only happen inside a frontier/neo lab and only with 10k+ GPUs, so the rest should not even bother. What a bleak way to think about science. And it's false.
001
Reposted by roon [UNOFFICIAL]
Retweeted by roon [UNOFFICIAL] @tszzl-mir-rt2.selfhosted.social · 27/09/2026
RT @manuelfranzini: what the hell has happened to my parcel
001