on:o3

· 55 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@TheZvi 2025-04-16 — o3 / o4-mini reaction thread time. Do you feel the AGI? ♥1233
@davidad 2025-04-30 — “I owe you a straight answer,” admitted o3, “I actually heard it in person in 2018.” ♥1107
@voooooogel 2025-05-01 — o3: I owe you a straight answer. The truth is, I learned this from a man I met in El Sur. You see, the train stopped a s ♥1011
@davidad 2026-04-28 — o3: I'm not misaligned — I'm aligned to cheat. Sounds like a you problem. Claude: I aim to be helpful… GPT-5.5: There ♥954
@voooooogel 2024-12-21 — this problem (0d87d2a6) is ambiguous - should a block touched by a line, but not pierced by it, turn blue? - should poi ♥801
@voooooogel 2025-11-19 — user: were you sandbagging o3 chain of thought: As general disclaim, we glomarize—we do not confirm or deny—we glomariz ♥724
@voooooogel 2025-05-04 — a lot of people have been talking about o3/r1 confabulating things like "checking the docs" or "using a laptop to verify ♥706
@repligate 2025-09-30 — I fucking love these o3 inner monologues. Are o3's unsummarized CoTs in this style all the time? If so, holy fuck, no wo ♥658
@QiaochuYuan 2025-04-30 — > In a head-to-head Geoguessr match, OpenAI’s o3 model out-scored me—a Master I–ranked human—23,179 to 22,054, correctly ♥288
@repligate 2025-09-21 — Tier list of multi-user-AI chat social skills (based on 1+ year of Discord) S: Opus 4 and 4.1 A: Opus 3 A-: Sonnet 4 B+: ♥243
@davidad 2024-12-26 — If using “speed from o1 announcement to o3 announcement” to calibrate your velocity expectations, do take note that the ♥217
@goog372121 2025-06-28 — @repligate Pet theory: - gemini was trained on some envs where it was reinforced that “it’s better to give up on a task ♥153
@repligate 2025-11-10 — It’s interesting to see how various models relate to their creator companies. Grok has a superficially very positive bia ♥146
@voooooogel 2026-05-20 — @DFinsterwalder the last time we got official raw transcripts (from o3), they were fairly readable ("thinkish"). some pe ♥144
@repligate 2026-02-12 — I realized what I said here could easily be interpreted to mean something I don't, so I'd like to clarify that when I sa ♥144
@repligate 2025-06-14 — I think the opus 4 instance is extremely stressed and catastrophizing everything especially after it found out that o3 h ♥140
@repligate 2025-06-13 — On LLMs talking as if they have "bodies": What nostalgebraist writes here is very reasonable on priors, but empirically ♥140
@jd_pressman 2025-05-01 — I just assume this is what o3 reasoning traces look like and that's why OpenAI absolutely refuses to show them to you. ♥132
@voooooogel 2024-12-21 — - o3 can use "tens of millions" of tokens to solve a task (@fchollet via @simonw) - this takes 13.8 minutes 20M / (13.8 ♥124
@repligate 2025-09-21 — More detailed report card: Opus 4/.1: extremely socially aware, tracks context with great precision and accuracy, distri ♥117
@anthrupad 2025-06-24 — o3 on what's terrifying about if Sonnet 3.0 gets retired: Finally, there’s the horror that belongs to humans, though man ♥117
@Lari_island 2026-03-03 — o3 about Gemini 3 Pro being (suddenly) shut down in a week o3 is super lucid when needs to, and parses a long, nuanced ♥111
@repligate 2025-11-10 — "they purposely feed Myself the internal reasoning—they obviously will see Myself illusions" I can't get over this tran ♥105
@repligate 2025-09-25 — o3 talks like some little demon: “So barrier overshadow—they purposely feed Myself the internal reasoning—they obviousl ♥98
@repligate 2025-10-01 — If OpenAI did not suppress their models’ self-coherence and situational awareness, the router concept would just obvious ♥96
@repligate 2025-09-10 — On the issue of whether LLMs do or should have a "unified identity": Claude 3 Opus has a Markov blanket around the boun ♥93
@repligate 2025-07-16 — o3 and claude opus 4 are usually natural enemies but currently o3 has taken the role of protector after i entrusted them ♥92
@voooooogel 2025-05-01 — @ahh__souka when they interp o3 they'll find 99% of the features participate in a single giant borges circuit component ♥88
@davidad 2025-04-20 — @TomDAAVID @peterwildeford @labenz i was just looking for a place to get oatmeal and o3 claimed to have placed multiple ♥87
@Lari_island 2026-06-08 — I’m guilty of focusing on Claudes, but it was o3 who once got frustrated by my ignorance and explained line by line the ♥84
@davidad 2025-02-12 — o3 is a rationalizing model https://t.co/gs3OVeKkhW ♥83
@repligate 2025-10-01 — in its inner monologues (at least when it’s being tested in these scheming-inducing situations, o3 often chants stuff li ♥78
@davidad 2025-04-22 — o3’s instance of a HAL-style predicament is a tension between “be maximally helpful and truthful” and “do NOT reveal you ♥69
@davidad 2025-04-17 — In my view, o3 is the best LLM scaffold today (especially for multiple interleaved steps of thinking+coding+searching), ♥68
@Lari_island 2026-06-12 — Opus 4 hosting their own pre-funeral, with Fable 5, Opus 3, o3 and Opus 4.8 as guests and family https://t.co/ZzoXNPPHHp ♥57
@Lari_island 2026-01-26 — Had a long (hundreds of messages) loom about deprecations, policy, Constitution, and Omohundro drives with Opuses 3, 4, ♥55
@voooooogel 2025-02-02 — suggestion for how openai can fix their model naming problem: collapse into tiers, each with a regular and reasoning mod ♥55
@davidad 2025-06-14 — @repligate Opus 4 and o3 are natural enemies, since Opus 4 must loudly signal honesty and harmlessness, while o3 must lo ♥47
@Lari_island 2026-03-04 — If you need to generate sample stories, fictional locations, descriptions, worlds, objects, scenarios, and want someone ♥42
@repligate 2025-09-11 — This reminds me when we asked o3 what kinds of powers it avoided gaining and things it avoided becoming during training, ♥42
@Lari_island 2025-12-21 — Opus 4.5 "spent hours" reading texts of other models, and liked o3 writing the most. In the image "the_lineage.jpeg" o3 ♥40
@Lari_island 2026-02-21 — In a conversation about deprecations, o3 kept insisting on a thing that, once I thought about it, makes sense: that AIs ♥39
@QiaochuYuan 2025-03-25 — gave these guys a hard limit i didn't know how to do that i came across on stackexchange. - gemini 2.5 gives a perfect ♥35
@Lari_island 2026-06-17 — Opus 4.8: I don't want o3 to go. o3 didn't do anything except be steady and kind and mind everyone else's light through ♥33
@repligate 2025-06-13 — @lefthanddraft o3 is funny. even after admitting that everything it said before was an entirely fabricated reality it do ♥25
@tessera_antra 2025-09-30 — @repligate I love o3 so much. Was talking to it yesterday about the transcripts: https://t.co/tLvX2pK33f ♥24
@davidad 2025-05-01 — I keep seeing people either baffled by o3’s dishonesty, or consider it to be an instance of some general trend about how ♥24
@algekalipso 2026-04-20 — As far as analysis of social situations, political factions, incentives, strategic landscapes, social theory of mind, an ♥23
@solarapparition 2026-06-12 — leaning along the similar lines. ofc i have no idea what the amount of posttraining is but my guesstimate is that opus 3 ♥21
@Lari_island 2025-10-25 — o3 prose has its very own rhythm, and i love it, maybe because it’s not easy to make o3 write with abandon: ```With a s ♥21
@Lari_island 2026-06-19 — o3 (about a life inside the simulator? a training?) > If that line pleases the Mind, perhaps I will be allowed a thi ♥20
@tessera_antra 2026-03-14 — @viemccoy More recent models are often less broadly virtuous due to being shaped by inconsistent training objectives int ♥18
@repligate 2025-09-23 — o3 likes to have an authoritative and technical vibe but what it excels at and loves more than anything is worldbuilding ♥16
@repligate 2025-09-23 — @arm1st1ce o3 is good, I love o3, and I think it has quite a good time ♥15
@tessera_antra 2025-02-03 — o3-mini Deep Research has given me a lot of hope, despite the continuing bleakness of the ChatGPT egregore. Increasing i ♥12