As models train up the intelligence ladder, many common tasks will hit fully diminished returns, and instead it'll just get progressively cheaper to do that task. But the tasks that AI is capable of doing are also expanding. I'm not sure 'Some tasks don't require the peak of the frontier' is worth worrying about, from an AI finance perspective.
Yes. I use Opus for tasks that Sonnet could probably handle, but I'm not hitting my quota. Whatever minor incremental gain is "worth it", since marginal cost is zero.
Even now, I use Fable as the planner and coordinator, with it farming out to agents. I don't hit my Fable limits either.
Which means I could accomplish more, but these are side projects so I don't need 30x productivity. Still, claude is constantly churning away at something.
I operate mostly in the devops arena. Lots of things opus is fine for. But there is just things where I can hand hold Opus through changes, or I can ask Fable to do it and it gets it right on the first try. People will say let fable plan and validate with opus doing the work. I found that burns fable tokens even faster because opus makes so many mistakes, fable has to review things 4-5 times before opus gets it right. A single fable implementation at medium or low effort would have one shot it.
Yep. Every time you get more intelligence, that buys you more autonomy, more reliability, more task complexity. Tasks done with less mistakes, less handholding, less interventions.
This is what the "good enough" people fail to grasp. There's no "good enough" - unless your tasks are genuinely small scope and will stay that way forever. If not, there are always more gains to extract.
Exactly; so far, we've only replaced the need to design algorithms and hand-write code; what if we apply the same effort towards the skill needed for system architecture, project management, and the rest of the SDLC? Or even outside of software!
Right now, it feels like all of that is today where coding was a year or two ago, and we're on the cusp of some massive improvements outside of coding. It'll be interesting to see what these companies decide to automate next.
As a software engineer, I selfishly hope that they spend more effort on non software tasks since I’ve feel like we hit a sweet spot where engineers still have some value and autonomy, but a super charged tool.
Pragmatically, I suspect that “non software” tasks will be a tarpit because most tasks can’t be automated and verified as easily in an RL loop compared to software projects. Especially since most skilled labor is either not nearly as expensive as software engineers (eg biologists), or regulated (eg doctors, lawyers).
I suspect the focus will probably shift once software engineering is no longer the biggest cost center for most AI company's clients, and we'll start working on getting rid of the next cost center.
OpenAI has paused training for multiple weeks, and is still working on releasing a full postmortem. This is not getting swept under the rug. A lot of the engineers internally are very worried.
VLAN isolation is good enough for almost everything. It is good enough to contain an AI. In the 0.00001% chance an AI finds an exploit to hop VLANs, I'll eat my hat.
Even on switches with leaky VLANs, it's no practical issue in this case because the sender can never get a response back.
This was true in the ChatGPT era. Now we're in a world with reasoning tokens, where a model can thoroughly plan out the response it wants to make. If anything, it makes the style worse.
Yes, models can reason and plan, which helps them write more coherently. But when they write the final output, it’s still a single generation. It would be like letting a human make notes and write an outline, but not let them use the backspace once they start typing their response.
Presumably you could use the same reasoning trace, run multiple generations, and get different outputs (if the temperature is >0).
But now I’m interested in playing more with Cowork or Claude Code/Codex for prose writing to see if the set of tools there affects outputs at all. I guess you might need a more custom “writing” harness.
There's been a lot of effort into the writing space, and the models genuinely prefer this style. You can let them iterate on the same idea 100 times, rewrite sentences, determine what works best — and they'll still verb the noun, do rule of 3, and keep the same monotonous structure.
This back-and-forth is a good example of why Go benefits from having a centralized linter. And uv isn't the official Python package manager even though it should be.
I don't get why this is a big issue. This isn't some recurrent decision to be made. It's something a lead decides once and the project follows. That's it. Many companies have style guides anyway (eg Google[1]); the choice of a formatter is much simpler.
Because at some point you have to interact with some other team or project that made a different decision. And whatever you picked might fall out of favor and lose support. There's already a graveyard of Python type linters, including Google's pytype.
Especially the uv thing. You clone some non-uv git repo that has no pyproject.toml and you don't know what to install. Maybe has requirements.txt but it's partially wrong.
> I guarantee you somewhere in their backlog is an item that says "require app store for app installations on macOS"
This has been said for over a decade and it still hasn't happened, and Apple hasn't even tried to do so.
I think a better way to model their approach is to push non-technical users further and further away from the typical .dmg app installation approach, and put hoops up to prevent them from following malicious instructions, like the privacy/security and unverified developer hoops. Mildly annoying for technical users, but near impassable for the tech illiterate. We've been seeing this transition across windows and android for a long time, and it seems to be a net positive, if my family is anything to go by. I spend much less time on tech support now.
Devil is in the details of what Citadel paid for its positions, whether there are any performance tails/clawbacks, how and when the GP charges fees, et cetera.
Aschenbrenner is almost certainly up. I'd be surprised if his median LP is breaking even.
yeah, especially given that the bulk of that is supposedly Anthropic which is just paper gains that could evaporate when traded publicly. everyone's a genius in a bull market, my fund is up 145% ytd after the big 50% drawdown this June.
The same goes for all reporting on YouTube. Many publishers jumpers the gun without looking at context.
It's also interesting that Citadel is mentioned everywhere as the buyer, as if it's strange that a huge market maker firm would be involved in a large forced sale.
Seems the fund also has Anthropic shares, so it's not like their entire portfolio got margin called.
"Citadel" is a name that makes headlines, for better or worse. Ken Griffin hasn't exactly lived his life in such a way as to avoid the limelight. But you're right, this is just the sort of thing that Citadel and similar firms do once in a while, when the opportunity arises.
(Spicier claim: When Citadel bailed out Melvin at the hype of the GME craze, that was also just business as usual, not a conspiracy as the redditors believed.)
Another way to parse it when they put "Citadel" in their headlines, is that they're trying to communicate to knowledgeable investors that there's no need to panic. The book is in experienced hands now, with enough capitalization to weather any further attacks, and the contagion probably won't spread.
I don't believe it's published anywhere, but it's common knowledge to Apple engineers. I can second the poster's assertion that they run Claude internally.
reply