Wow what an absolute nothingburger of an article.

Someone at the Pentagon may have used Claude (maybe directly or maybe indirectly through a Palantir tool), to do something that may have had something to do with a really large military operation that involved possibly tens of thousands of people. Maybe. We don’t know. Did I say maybe?

Amazing. The rest of the article reads like filler from a college freshman when you’re required to meet a word count quota. A bunch of random factoids strung together.

I bet they used Claude to write this stuff.

> The Indian consumer market.

You've said this a couple times in this thread now. Do you have any evidence that most of his audience is in India, to make that claim that his ethnicity matters?

Does this mean they’re struggling financially?

Yet more disruption caused by coding agents, I’m sure. We saw it quite visibly with Tailwind, now I can see if code editors are maybe struggling too, especially something like Zed which was probably still used mostly by early adopter type People, who have early adopted TUI coding agents instead.

I only use cursor and zed to browse code now.

I don't think it means they're struggling financially. I think it means they're not steering the ship alone any more, and are responsible to others. That's how accepting investment money generally works.

Ironic, because AI Agent integration is their business model.

Now it is, after their pivot from remote pair programming via collaborative text editing, not originally.

For now. And also largely because it's easier to get that up and running than the alternative.

Eventually, as we ramp up on domestic solar production, (and even if we get rid of solar tariffs for a short period of time maybe?), the numbers will make them switch to renewable energy.

Musk has been talking about this (generalizing the self driving model for their Optimus bot) for a while now.

Which is why their strategy (purely vision/photons in, controls out) seems to be more widely applicable and scalable over time.

And waymo seems to be arriving there too as they keep reducing the equipment (it would seem)

Just wanted to echo that my experience is almost identical (production elixir/phoenix for 10 years), and also very rarely touch erlang. But it does leak in every now and then.

And it'll feel less intimidating once you code in elixir, but I still find it more cryptic and less approachable the 3-4 times I've had to splunk into erlang to debug something in the last decade.

I'm starting to wonder if people doing what were previously unconventional workflows (which may not be performance optimized) are affecting things.

For example, today, I had claude basically prune all merged branches from a repo that's had 8 years of commits in it. It found and deleted 420 branches that were merged but not deleted.

Deleting 420 branches at once is probably the kind of long tail workflow that was not worth optimizing in the past, right? But I'm sure devs are doing this sort of housekeeping often now, whereas in the past, we just never would've made the time to do so.

I didn’t quite understand why they were randomly giving people $50 in credits. But I think this is why?

no, it’s for Max subscribers to enable “use API when running out of session limit”. the assumption (probably) being that many will forget to turn it off, and they’ll earn it back that way.

This was my first thought, but by default, you have no automatic reload of your prepaid account. Which I think is for once user friendly. They could have applied a dark pattern here.

Wow, I have been using Open 4.6 and for the last 15 minutes, and it's already made two extremely stupid mistakes... like misunderstanding basic instructions and editing the file in a very silly, basic way. Pretty bad. Never seen this with any model before.

The one bone I'll throw it was that I was asking it to edit its own MCP configs. So maybe it got thoroughly confused?

I dunno what's going on, I'm going to give it the night. It makes no sense whatsoever.

I am also _not_ happy. I tried the `/model` command and I could not switch back to Opus 4.5. However, the command line option did let me set Opus 4.5:

``` claude --model claude-opus-4-5-20251101 ```

I will probably work with Opus 4.5 tomorrow to get some work done and maybe try 4.6 again later.

It was better today. I dunno if there was a regression in a corresponding cc version that was maybe quickly patched?

It felt like it was at least back to opus 4.5 levels.

To me its obvious.

Theres a trade off going on - in order to handle more nuance/subtleties, the models are more likely to be wrong in their outputs and need more steering. This is why personally my use of them has reduced dramatically for what I do.

I'm not sure. Anthropic probably feels safe enough for the foreseeable future given that they've been focused heavily on business customers (cowork, claude code) who just pay straight up for the service.

Whereas OpenAI seems to be huge in the consumer market (where downward pressure on pricing make ads more likely). They're trying with Codex but all the other stuff (the legal, financial, tooling with cowork, etc) seem to be a lot more fleshed out on the Claude side.

So they can probably get away with this for a while. That's my guess though.