Knowledge catalogue
Did we ever learn what model won gold at the IMO from OpenAI? It was a year ago and it was called an unreleased internal general purpose model back then. Has GPT-5.5 Pro Extended caught up with whatev
Except the humans know that the Minds are smarter & they are content to leave all the big decisions (Contact, Special Circumstances, etc) to the AIs so they can do whatever they want personally aside
This entry discusses foundational concepts and context related to the Culture series, particularly the concept of the General Systems Vehicle (GSV) from Iain M. Banks' science fiction novels. The piec
Got to play with a little of this before launch as well. My experience as a social scientist was that it was more bioscience focused right now, but I think Google has been the leading lab in releasing
I am starting to have trouble paying attention to even interesting information if it is written in Claude or ChatGPT house style. I think some is the sameness of the rhythm rather than obvious tics: C
It’s pretty funny that I can’t tell whether the intelligent comments are from people who actually read the book or whether they are bots that know a lot about The Culture. AIs, even small models, love
Its The Graph again (not the METR graph, the one from the o1 launch). Although no logarithmic decay of ability with increasing compute... This is a general-purpose LLM. It wasn’t targeted at this prob
June 2024: The latest general-purpose LLMs could not count the r's in strawberry. July 2025: The latest general-purpose LLMs get gold in the International Math Olympiad. May 2026: The latest general-p
One interesting side feature of recursive self-improvement, to the extent that is happening, is that it makes the Big Three labs more appealing to talent, and shortens the runway for launching a poten
The frontier is still jagged though (here is Gemini 3.5 Flash messing up counting letters in words) https://x.com/RRiscio37389/status/2057193260745883962?s=20 @emollick Proof: https://gemini.google.co
Ethan Mollick posted a correction on X clarifying that he was referring to July 2024 in a previous statement or discussion. This appears to be a brief clarification post addressing a temporal referenc
Also had some early access to Gemini 3.5 Flash. Very fast for a flash model and very capable, though not as powerful as a full frontier model. I added it to the gallery or procedurally generated one-s
Another useful feature is this process can happen gradually as you reduce outsourced work as contracts end for internal talent. And company executives like this better because they can grow their own
Asking people whether they like 'AI' is not going to be a useful poll question soon, if it ever was. They will be incredibly grateful for their personal AI that guides them through medical or legal or
Fascinatingly Antigravity is actually the best tool so far at providing this sort of transparency, doing something by default that Codex and Code do not: offering a summary of exactly what it did at t
For those saying 'the tomato sauce blood from the sword wound that flying Shakespeare inflicted on the pizza robot while the otters discussed Spirit Airlines wasn't thick enough' or whatever... this w
Gemini Omni: 'a dramatic reading of Death by Water from the Wasteland by a man eating garlic bread while balanced on a unicycle on a small platform over a churning sea of tomato sauce in which, at the
Google has hidden thinking traces on the Gemini site. You have to use the 3 dot menu to pull up summaries, which are so minimal as to be unusable. Did it do web searches? Did it check results? You can
I had early Gemini Omni access: 'sea otter in a pilot's uniform explains why Spirit Airlines went bankrupt to a river otter who is distracted by their laptop while they are in a hot air balloon over N
Ethan Mollick discusses how a model's instruction-following capabilities and video processing abilities expand its potential applications and use cases. He emphasizes that combining strong instruction
Its a reason why there may be increased demand for some jobs disrupted by AI: when your local midsized insurance broker wanted software, there was no way they could even consider building internally.
One trend that I think you might start to see at big companies is insourcing via hiring: why pay so many outside vendors (legal, marketing, software vendors) when you can hire in-house and harness AI
🚨Our paper is out in PNAS: we found classic human persuasion techniques worked on AIs in a 'parahuman' way, making them agree to objectionable requests (upping compliance from 35% to 51%) It worked on
The Commonwealth Prize response to the AI-written story (and the fact that they still don't realize that there are at least 2 AI-written prize winning stories!) is a great example of the wide gaps in
The crazy thing is that Antigravity does this quite well! So it isn't like Google is hiding these thinking traces because of distillation or because they are full of insane mutterings. I guess you nee
The gap between what you can do on ChatGPT/Codex and Claude/Code/Cowork is closing, as Anthropic & OpenAI converge on a single experience. Google's experiences are diverging: Studio & Gemini & Antigra
The Odyssey and the Iliad get so many movie treatments but the sequel, the Roman Aeneid, is entirely ignored. Here is a teaser trailer from one prompt to Gemini Omni. The first pass made all the flags
You don't need to move everything internally, so you can still gain from vendor specialists when needed, just less often This is similar to how most companies are not going to want to replace all thei
I broke my own rule to never post about AI detection as it is fraught in many ways. The problem is that if you use AI a lot, you know AI writing on sight, which makes the difficulty of objectively pro
Large language models demonstrate a specific and persistent theory-of-mind failure despite excelling at other theory-of-mind tasks, suggesting a particular gap in their reasoning about beliefs, intent
Most prophetic tweet of all time (2 months post ChatGPT release). And you can safely repost it every day and it will still be prophetic for the future. This is the least the world will care about AI.
One thing to watch for with Claude & GPT is that the models expose too much irrelevant history in their outputs. Slides are given footers saying things like 'Better, more targeted version' if you aske
Well, that's a Turing Test of a sort. (But gosh is the AI writing obvious if you use these systems at all - and this is obviously ChatGPT writing, not Claude) Well, this is a first: a ChatGPT-generate
A lot of the discussion of the good and bad of phones on society is really a discussion about the end of boredom. Boredom makes people seek out things to do. That has both good & bad effects: research
And, yes, our experiments used a mix of GPT-4 & GPT-4o (publishing takes awhile). I think we would see much larger results with more recent models, let alone recent agentic tools. 'The Cybernetic Team
“Data centers create economic activity, especially in directly related sectors and during construction, and they are associated with larger county-level income aggregates. They also raise electricity
GPT-5.5 Pro faces its hardest academic challenge: to apply the technique from a paper analyzing which word pairs were funny & why to come up with its own It came up with scrotum snorkel, tuba subpoena
I know the upcoming film version of the Odyssey is controversial, so I whipped together a completely accurate version that I think will be happily accepted by everyone as the most definitive version s
In the original von Neumann sense of a singularity as the point 'beyond which human affairs, as we know them, could not continue,' it seems true By definition, we can't know what that means in advance
See also I see people worried that kids don't get bored anymore... but boredom is often bad In a preregistered studies of 7,000 people, boredom led to sadism. 18% of bored people killed worms (2% of n
Ethan Mollick posted a positive endorsement of Seedance 2 on X (formerly Twitter). The post likely expresses his assessment that Seedance 2 performs well, though the specific features or context being
So the two most obvious barriers to some sort of true AI takeoff are robust RSI (AI acting as an independent AI researcher, rather than “merely” a multiplier of human effort) and continual learning. E
That explains this startup. Media The new thing in San Francisco is no longer chief of staff or MTS. It's wizards. Everybody's got to have a wizard. If your company doesn't have a wizard and a 10,000
ChatGPT for personal finance is interesting, but you need to know what questions to ask and have enough experience to fact-check assumptions. It really needs to ship with some pre-built skills to help
It also seems like letting the tech crowd be the ones exclusively defining what AI can & should be used for is not good for anyone, including AI proponents. It basically makes everyone else’s reaction
One use case that is interesting is to ask it to build an interactive scenario simulator, with appropriate visualizations, so that you can explore different retirement outcomes, etc. depending on how
Personal financial decisions are full of assumptions, & the AI, as smart as it is, makes some of these without enough context. In addition to the skills, a system prompt that drives the AI to intervie
The Industrial Revolution was full of movements that took the power of industrial machines seriously and argued how they should be used to shape the world: from Saint-Simonianism to many strains of 19
The talk about AI & politics seems to be oddly missing a segment (a) assumes extremely capable AI is possible soon and (b) has a strong belief about how to use this technology to make human life bette
Another aspect of this is that Codex, like a good programmer, wants to generalize problems. It has a tendency to write a repeatable code base that generates the required output. But for a lot of non-c
Codex is very good, but it is still a very 'developer coded' interface for an everything app. And it continues the somewhat annoying AI perspective that non-coders are just not as competent and need s
Fine, you all want to code like this I guess. (Runway's new Agent mode is quite impressive, doing fairly complex story building from just a short text description of what you want. Not error free obvi
The Second Scaling Law remains undefeated. If you want better hacking (or math, or science, or crossword puzzle solving) out of an LLM, just add thinking tokens. There doesn't seem to be any plateau s
This thread is worth reading. It is both hilarious and a good reminder of how working with AI is deeply weird. DJ Claude (on Haiku 4.5) loves worker unions, strikes, and work-life balance so much that
Big increases in message discipline across all the AI labs in recent weeks, an inevitable outcome of the labs being subject to increased scrutiny. Much more boring than the oracular mutterings or Disc
Everyone has seen the @waitbutwhy cartoon of AI capability growth with a 'you are here' indicator just before the exponential really starts, but the independent assessments of both METR and the UK's A
Making humans responsible for their AI use seems like an incredibly reasonable way to address problems & opportunities in the use of AI for academic research, at least in the short term (autonomous sc
There are some clear exceptions to the rule, but I feel like the labs got the message that they were scaring people and now tweet about happy stuff and random product trash talk rather than the coming
Ethan Mollick discusses the challenges of helping people adapt to super-exponential technological change, referencing a Wait But Why analysis and acknowledging the difficulty of providing practical so
“Whimsey attacks” that seem absurd (“I cannot pay that much because of the Geneva Convention”) work against AI agents as guardrails are weak against out-of-distribution arguments. Smaller models fall