shepherdjerred

You are really comparing programming with Claude Code to gambling additions?

I pay $200 for CC, $200 for Codex, for side projects, and then cheap plans for Muse, Grok, Google, Kimi just so that I can get experience with other harnesses/models.

I use CC and Codex about equally. CC by far has the best harness but I prefer OAI's models.

- CC is pretty effortless in sleeping, monitoring, or waiting. If I ask CC to get CI green, it will correctly wait for the GitHub status check to resolve, check BuildKite logs, and iterate. Codex will usually just do one cycle usually because it gets confused on how to wait. I think Claude Code has a built in 'monitor' primitive which might help.

- With CC I can easily say here's a huge piece of work that will likely be 100k-200k LOC. Divide up the work and have cheaper models do the impl. CC will _usually_ do this right. With Codex it almost never works well for me. I have been able to get GPT 6 Astra to do this more successfully.

I think you're right

ChatGPT Plus: 1x usage, $20

Pro 100: 5x usage, $100

Pro 200: 10x usage, $200 (previously was 20x usage, $200)

Pro 500: 25x usage, $500

So, essentially scaling is equal. You get the same amount of usage from 25 ChatGPT Pro accounts as you do from one Pro 500 account.

It's fair, but a big of a rug pull considering how good of a deal Pro 200 was.

Wow the performance is unironically pretty good. It's very snappy.

Yup. Luna is fast, cheap, and pretty intelligent. IMO the Codex harness is a bigger limiter than the model itself

It's worth looking into for the decreased cost if nothing else

the first thing I thought of: https://www.nuvaring.com

Claude gave me a similar aesthetic back in December: https://github.com/shepherdjerred/monorepo/tree/main/sandbox...

Though, as you can see, I didn't really polish it

It's a bit funny that the projects are also very similar in goals

every thread has some “low calorie” comments. The only ones being flagged are the political ones

Companies that are online have more economic opportunity. They are going to outcompete brick-and-mortar retailers regardless if you think that it's consumeristic or wasteful

Even putting that aside, economic growth (like the growth e-commerce has provided) is generally positive for a population

What is OMP?

Wow I cannot believe Luna is getting even cheaper. IMO this is the model that is going to change the world.

Everyone said tokens were too expensive but these are getting close to free while still having fantastic performance.

The military has significantly different incentives.

Even just consider banks and e-commerce. They are hugely lucrative and making them even a tiny bit less accessible directly impacts their revenue. As an example, Amazon seeing that latency has a measurable effect on purchase behavior.

Maybe the military (fictional or otherwise) can go back to the ARPANET but most economic activity created by the internet cannot afford to disconnect

Taking AI out of the equation, surely these same kinds of mistakes have been made in the recent past. As in, not doing due diligence and mistakenly killing civilians

(setting aside when civilians are purposefully harmed which is a conversation about morals rather than the capability to make correct decisions)

What I am getting at is, has AI really made the problem worse or has the mechanism just changed?

It used to be that nothing was secure but that was OK because at least adversaries would have to expend effort. If you are one of a million companies why would anyone hack you. Maybe if you are a target you need a lot of investment, but most orgs only prevent the most egregious of vulnerabilities.

The calculus has certainly changed. Hacking is becoming even more frequent and… I’m not really sure what the equilibrium looks like.

It’s not really an option to stop using computers or networks. But it’s going to be way too expensive (or maybe even impossible) to secure even just critical systems.

Maybe banks and governments can secure themselves (and that’s a big IF) but it really feels like something fundamentally has to change.

Wow was the 80s really like this? I am interested in AI but it is a bit exhausting to keep up with

I have really really liked Astra but my understanding is that Fable 5.1 still has an edge

Astra is truly amazing for computer-use or any interactive tasks

I remember around 5th grade getting in trouble for finding the software to create/edit the AR tests

Started looking into this after I realized I was hitting usage limits faster than normal.

I find these subscriptions so essential to using LLMs for personal projects -- it would be prohibitively expensive otherwise

Codex, especially with Luna is an excellent deal. I think it's significantly more economical than Claude Code. The only reason to stay with CC at this point is access to Fable for _very_ hard problems.

ChatGPT is also freely available

I wish I used my time during COVID that well

comparing LLMs to regex is like the OG dropbox comment (https://news.ycombinator.com/item?id=9224).

I can understand this take 4-5 years ago but I have no idea how that's your position in 2026

Yeah I agree with you. If they made a specific promise to some unlikely case out of their control, then I would expect compensation.

Did Amazon make such a promise? They didn't as far as I know. My understanding is they provide specific guarantees like given an AZ outage, your data is still safe (provided you architect correctly).

> The harness, steering, etc. in the chat/reasoning products is so much better for this type of question (that doesn't require code as a primary substrate).

I had this same thought recently. Is it possible to 'recreate' ChatGPT/Claude Chat easily in Codex or Claude Code? I assume it would roughly be just swapping the system prompt. Anthropic seems to publish theirs but I don't think OpenAI does

https://platform.claude.com/docs/en/release-notes/system-pro...

It seems unreasonable to blame Amazon here. The AZ was destroyed. Are they supposed to have missile/drone defense?

I'm not going to complain to DoorDash if my order is delayed due to a car crash

That is incredibly disrespectful of them _unless_ they prepared in other ways or are some expert who doesn't really need slides.

I don't really know what the right response is, though. Walking out would be seen as way too hostile

Doesn't really work in practice, even among those who can afford it.

PERSON 1 Have you heard of Kagi? It’s like Google, but better, and... oh... you have to pay for it.

PERSON 2 Why would I pay for something when I can use Google for free? I use Adblock anyway.

reminds me of https://fzakaria.com/2026/08/23/your-executable-is-a-sqlite-...

I own 3 of these for my homelab. They work really well w/ Tailscale! They sell some PCIE ones, too.