@JuliusCasio
Account based inGermany
About this account
- Account based in
- Germany
- Connected via
- Germany App Store
Account-level information from X, not a live location or the device used for a specific post.
interested in: ai, mech int, writing, cycling, fitness, defense, biotech, europe
munich
Joined October 2022
- Tweets2K
- Following2.3K
- Followers242
- Likes34.7K
a bit late … by now almost everyone sounds like ChatGPT
Astra is great at adapting your writing, so we launched Writing Styles to help ChatGPT sound like you
I use it all the time for slack messages and quick docs
Takes less than 30 seconds to set up! chatgpt.com/?surface=work&wr…
Navier this, Stokes that - when will we solve helmets though?
This model represents a step-function improvement on many benchmarks, and its training is ongoing.
Our internal model group arrived at the Navier–Stokes solution in 88 hours, using around 10,000 coordinating AI agents.
Throughout the effort, we maintained the strict safeguards—including monitoring and isolation—that we apply to all our frontier evaluations.
very cool! Also, I feel so comput- and intelligence-poor. 10.000 agents! Imagine, as a country, you are still not taking the implications of this rapid progress seriously.
This model represents a step-function improvement on many benchmarks, and its training is ongoing.
Our internal model group arrived at the Navier–Stokes solution in 88 hours, using around 10,000 coordinating AI agents.
Throughout the effort, we maintained the strict safeguards—including monitoring and isolation—that we apply to all our frontier evaluations.
so there we have it. And remember it’s only going to get faster from here on
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
Such a good point! Made me realize how little difference it'd make if some accounts are ai or not. 90% of your feed you will never see irl! Much of "social" media is already parasocial in a way where replacing one side with AI wouldn't change much about your relation to them
Props to Anthropic because their resets are REAL resets: whenever codex resets, your weekly usage reset gets updated too to happen in seven days, eventhough before it was tomorrow. This Claude reset does not do this.
Day 2 without Codex.
I visited a physics lab of a friend.
Spent two hours on the bike. Weather was really good. Someone said there was a banked reset, but I’m saving that to spend it on Astra.
is the caveman skill github.com/juliusbrussee/cav… really as good as 102k stars suggest?
Day 1 of life without Codex
I bought twenty dollars of credits. They were gone after 25 minutes. Will never do this again. Then I left the office.
Fynn retweeted
There are two ways to escape your life: avoid the future, or avoid the present.
Normal people tend to fall into the first category. Gen Z doesn’t believe in the future anymore and prefers a never-ending present with no long-term objective.
Startup people tend to fall into the second category, but risk being trapped in permanent deferral - living in a constant liminal space.
Finding how to reconcile both is where happiness resides.
Fynn retweeted
I want to prevent a race into unmonitorability kicked off by confused reporting. The depth of the computation graph for our present frontier models, including Astra, is within a factor of two of GPT-4.
OpenAI has worked to preserve and utilize chain-of-thought monitoring since our very first reasoning models. We deeply care about this technique, as it can give us a view into how model alignment generalizes from its training distribution. I do think it is fragile and unfortunately trending in a negative direction, for reasons not contingent on architecture changes that I will write about soon. But there are things we can do to strengthen it, and it's a core goal of our current research program.