The biggest problem with satellite internet was the costs involved, which SpaceX has pretty much solved.
Datacenters in space, on the other hand, are a terrible idea because of the laws of physics, which will not get "solved" anytime soon. But don't take it from me, listen to this guy with a PhD in space electronics who worked at NASA and Google:
Check the authors history. They are both anti AI and anti Elon. I think I feel a lot more confident staying optimistic and assuming that the SpaceX and xAI team have done their research about this. I know a lot of people are heavily biased in this matter due to politically not liking Elon or not liking AI, but I also think it's fair to say these companies have many very smart individuals working for them. If they have come to the conclusion that this is viable, then I have much more faith in what they are saying over one guys opinion who is biased against them and saying it's a bad idea.
You're also passing these judgements without knowing their full plan. Maybe we only know one part of the plan and maybe other details have not been announced. They may have a much bigger plan for this than just the specific information we have.
Having previously criticized someone doesn't make your technical analysis biased. It just means you noticed similar problems previously. Conversely, "I used to support him so I'm not biased" is given unearned credibility when really it just means you were late to noticing the obvious.
Technical analysis most definitely can be biased due to political leanings. This is why there is the whole idea that research can often be bought and paid for to get the results you desire. Because they are biased with money. Certain ideas or theories of how things could be done could very easily be overlooked or excluded by someone trying to dig for reasons to say something won't work.
What I am saying is that clearly SpaceX/xAI feel that this is a viable option based on many experts research/facts that are more knowledgeable than a single bloggers opinion. If I am thinking rationally why would I choose to believe a single random person over a group of experts banking A LOT of money that they have a solution that works?
You are arguing against something I didn't say. I never claimed bias doesn't exist. My point is that having previously criticized someone is not evidence of bias. You are treating "this person has been critical before" as inherently discrediting, when it is just as likely they were right before and are right again now. Conversely, "I used to support him so I am not biased" is given unearned credibility when really it just means you were late to noticing the obvious, or got it wrong previously.
As for dismissing the article: the author has a PhD in space electronics, worked at NASA, and spent a decade at Google including on AI capacity deployment. He walks through power, thermal, radiation, and communications constraints with actual numbers. You do not get to hand-wave that away with "he is anti-Elon" and then defer to "the team spending the most money." That is not rational analysis, that is fandom.
And the idea that SpaceX's experts looked at this and concluded the combination makes strategic sense - seriously? This is the same playbook Musk has run repeatedly: SolarCity into Tesla, X into xAI, now xAI into SpaceX. Every time there is a struggling asset that needs a lifeline, it gets folded into a healthier entity with Musk negotiating on both sides. xAI is burning $1B/month. There is already a fiduciary duty lawsuit over Tesla's $2B investment in xAI. The "space data centers" rationale is a pretext for giving xAI investors an exit through SpaceX's upcoming IPO. This is not a strategic vision, it is financial engineering solving an obvious problem for Elon.
Meanwhile, Grok has been generating sexualized images of children, the California AG has opened a formal investigation, the UK Internet Watch Foundation found CSAM attributed to Grok on the dark web, Musk personally pushed to loosen Grok's safety restrictions after which three safety team members quit, and xAI's response to press inquiries was the auto-reply "Legacy Media Lies." This is the company whose judgment you are trusting over a domain expert's detailed technical analysis.
I guess you'll have to wait and see what ideas they have to deal with this. If they can't manage the heat they aren't going to spend billions launching these things just for fun.
Which is precisely why I said originally that data centers in space have never been a thing and will never be a thing. Because the whole premise is "it's cold in space so that's great for data centers", but that fundamental premise is fundamentally wrong and based in a misunderstanding of the physics involved. There is no other redeeming argument for it, therefore it's not going to happen. Anyone trying to sell you on data centers in space is grifting.
I'm almost certain it will be significantly worse.
The Excel sheet will have been tuned over the years by people who knew exactly what it was doing and fixed countless bugs along the way.
The Claude Code copy will be a simulacrum that may behave the same way with some inputs, but is likely to get many of edge cases wrong, and, when you're talking about 30 sheets of Excel, there will be many, many of these sharp edges.
I won't disagree - I suffered from insufficient damning praise in my last sentence above.
IMHO, earned through years of bleeding eyeballs, the first will be riddled with subtle edge cases curiously patched and fettled such that it'll limp through to the desired goal .. mostly.
The automated AI assisted transcoding will be ... interesting.
My assumption is that with the right approach you can create a much much better and reliable program using only Claude code. You are referring to yolo coding results
> I helped one recently almost one-shot[3] converting a 30 sheet mind numbingly complicated Excel financial model to Python with Claude Code.
I'm sure Claude Code will happily one-shot that conversion. It's also virtually guaranteed to have messed up vital parts of the original logic in the process.
It depends on how easily testable the Excel is. If Claude has the ability to run both the Excel and the Python with different inputs, and check the outputs, it's stunningly likely to be able to one-shot it.
Something being simultaneously described as a "30 sheet, mind-numbingly complex Excel model" and "testable" seems somewhat unlikely, even before we get into whether Claude will be able to test such a thing before it runs into context length issues. I've seen Claude hallucinate running test suites before.
>I've seen Claude hallucinate running test suites before.
This reminded of something that happened to me last year. Not Claude (I think it was GPT 4.0 maybe?), but I had it running in VS Code's Copilot and asked it to fix a bug then add a test for the case.
Well, it kept failing to pass its own test, so on the third try, it sat there "thinking" for a moment, then finally spit out the command `echo "Test Passed!"`, executed it, read it from the terminal, and said it was done.
I was almost impressed by the gumption more than anything.
I've been using Claude Code with Opus 4.5 a lot the last several months and while it's amazingly capable it has a huge tendency to give up on tests. It will just decide that it can commit a failing test because "fixing it has been deferred" or "it's a pre-existing problem." It also knows that it can use `HUSKY=0 git commit ...` to bypass tests that are run in commit hooks. This is all with CLAUDE.md being very specific that every commit must have passing tests, lint, etc. I eventually had to add a Claude Code pre-command hook (which it can't bypass) to block it from running git commit if it isn't following the rules.
I haven't seen it bypass my hook yet (knock on wood). I have my hook script [0] tell that its commits are required to pass validation, maybe that helps push it in the right direction?
It compacted at least twice but continued with no real issues.
Anyway, please try it if you find it unbelievable. I didn't expect it to work FWIW like it did. Opus 4.5 is pretty amazing at long running tasks like this.
I think the skepticism here is that without tests or a _lot_ of manual QA how would you know that it did it correctly?
Maybe you did one or the other , but “nearly one-shotted” doesn’t tend to mean that.
Claude Code more than occasionally likes to make weird assumptions, and it’s well known that it hallucinates quite a bit more near the context length, and that compaction only partially helps this issue.
If you’re porting some formulas from one language to another, “correct” can be defined as “gets the same answers as before.” Assuming you can run both easily, this is easy to write a property test for.
Sure, maybe that’s just building something that’s bug-for-bug compatible, but it’s something Claude can work with.
I generally agree with you, but I tried to get it to modernize a fairly old SaaS codebase, and it couldn't. It had all the code right there, all it had to do was change a few lines, upgrade a few libraries, etc, but it kept getting lots of things wrong. The HTML was wrong, the CSS was completely missing, basic views wouldn't work, things like that.
I have no idea why it had so much trouble with this generally easy task. Bizarre.
where exactly have you seen excel forumalas to have tests?
I have, in my early careers, gone knee deep into Excel macros and worked on c# automation that will create excel sheet run excel macros on it and then save it without the macros.
in the entire process, I saw dozens of date time mistakes in VBA code, but no tests that would catch them...
And also - who understands the system now? Does anyone know Python at this shop? Is it someone’s implicit duty to now learn Python, or is the LLM now the de facto interface for modifying the system?
When shit hits the fan and execs need answers yesterday, will they jump to using the LLM to probabilistically make modifications to the system, or will they admit it was a mistake and pull Excel back up to deterministically make modifications the way they know how?
Alright so it is implicit that it is someone’s duty to learn Python. Don’t get me wrong, Python is better than Excel in a million ways, but not many companies are competent at hiring people who are competent at Python - including software companies from my personal experience hah.
I’ve also heard plenty of horror stories of bus factor employees leaving (or threatening to leave) behind an excel monstrosity and companies losing 6 months of sales, so maybe there’s a win for AI somewhere in there.
I'm having trouble reconciling "30 sheet mind numbingly complicated Excel financial model" and "Two or three prompts got it there, using plan mode to figure out the structure of the Excel sheet, then prompting to implement it. It even added unit tests to the Python model itself, which I was impressed with!"
"1 or 2 plan mode prompts" to fully describe a 30-sheet complicated doc suggests a massively higher level of granularity than Opus initial plans on existing codebases give me or a less-than-expected level of Excel craziness.
And the tooling harnesses have been telling the models to add testing to things they make for months now, so why's that impressive or suprising?
No it didn't make a giant plan of every detail. It made a plan of the core concepts and then when it was in implementation mode it kept checking the excel file to get more info. It took around ~30 mins in implementation mode to build it.
I was impressed because the prompt didn't ask it to do that. It doesn't normally add tests for me without asking, YMMV.
Did it build a test suite for the Excel side? A fuzzer or such?
It's the cross-concern interactions that still get me.
80% of what I think about these days when writing software is how to test more exhaustively without build times being absolute shit (and not necessarily actually being exhaustive anyway).
having worked in large financial institutions, this would be a step improvement
the largest independent derivatives broker in australia collapsed after it was discovered the board were using astrology and magicians to gamble with all the clients money
I have seen Excel used in places that manage millions of euros of other people's money.
I have seen Excel used for financial planning
I have seen Excel used for managing people's health data.
I have BUILT a test suite for a government offical use communication device - inside Excel. The original was a mish-mash of Excel formulas and VBA. I improved the VBA part of it by adding a web cam to the mix.
I don't sleep well at night knowing how many very very essential things are running on top of Excel sheets passed down like stories around a campfire.
I doubt that 30-sheet Excel model itself was validated by the most rigorous of standards. Unless it had its own test suite before the one-shot (which could be leveraged by the AI), people have probably been taking its outputs for granted for years.
My TL;DR is that they tried to run an on-device model to classify expenses, it didn't work even for simple cases ("Kasai Kitchin" -> "unknown"), they went deeeeeep down the rabbit hole to figure out why and concluded that inference on their particular model/phone is borked at the hardware level.
Whether you should do this on device is another story entirely.
> What's to be gained... by offloading inference to someone else?
Access to models that local hardware can't run. The kind of model that an iphone struggles to run is blown out of the water by most low end hosted models. Its the same reason that most devs opt for claude code, cursor, copilot, etc. instead of using hosted models for coding assistance.
Claude code produces stuff orders of magnitude more complicated than classifying expenses.
If the task can be run locally on hardware you own anyway, it should.
I would really not want to upload my expense data to some random cloud server, nope. On device is really a benefit even if it's not quite as comprehensive. And really in line with apple's privacy focus so it's very imaginable that many of their customers agree.
China would like to have a word with you. Soy milk in particular is hugely popular for breakfast, and there's about a zillion other ways to eat it too.
However, one of the big players in this space (Aerotel) nearly went belly up during COVID and cut their offerings drastically. They seem to be recovering though: https://www.myaerotel.com/en-uk
reply