What a sad path to see such a bright star going down. I guess it’s not a huge surprise, but it really does paint a bleak picture of technology to see how narrow the range of likely outcomes is. Doesn’t matter if you built the foundation for the future and cured cancer - the most likely outcome is being back to optimizing for engagement and revenue.
Like Anthropic? Due to Altman being a lying psychopath, most of the talent left OpenAI, which is now fighting for its life -- but now they can't claim to have moral high ground or the best researchers anymore; profitability's the only way out remaining to them.
Who knows? It could have always ended up this way anyway. But Altman had a pretty big role in summoning his own competition.
> "To Figma: I can see a world where this post does numbers in the Figma internal Slack. If that’s the case and you’re reading this from Figma: this wouldn’t have happened if you hired me last year when I was interviewing. Your loss, big dawg."
What a counterproductive way to end an otherwise good set of points. Gives the appearance of bitterness and a desire for vengeance.
From time to time I've tried using ChatGPT for financial modeling, and I have to say my experiences don't inspire much confidence.
Just this past week I used it to generate a simple model of a few different scenarios related to an investment property I own.
The first problem I ran into is that it was unable to output a downloadable XLS file. Not a huge deal - it suggested generating CSV tables I could copy/paste into a spreadsheet. The outputs it gave me included commas in a handful of numbers over 1,000 (but not all of them!) which of course shifted cells around when brought into Google Sheets. We pivoted our approach to TSV and solved this problem. Big deal? No. Seemingly basic oversight? Absolutely.
This is where the real fun began. Once I started to scrutinize and understand the model it built, I found incorrect references buried all over the place, some of which would have been extremely hard to spot. Here's my actual exchange with ChatGPT:
- - - - - - - - - -
> Can you check the reference in cell F3? It looks like it's calling back to the wrong cell on the inputs tab. Are there similarly incorrect references elsewhere?
> Yes, F3 is incorrect, and there are multiple other incorrect references elsewhere: (It listed about 30 bulleted incorrect references)
Bottom line -
- Many formulas point to the wrong Inputs row because of the blank lines
- The Sell + Condo section also has a structural design problem, not just bad references.
The cleanest fix is for me to regenerate the entire AnnualModel TSV with:
- all references corrected
- all 15 years included
- the condo scenario modeled properly with a separate housing asset column
- - - - - - - - - -
This was me asking about the exact output I had just received (not something I had made any changes to or reworked.)
There are plenty of domains where I have enough faith and error tolerance to use ChatGPT all day, but this just sends a chill down my spine. How many users are really going to proof every single formula? And if I need to scrutinize to that level of detail, what's the point in the first place?
All this is is evidence that Jack Dorsey had no idea what half the employees at Block were doing before he decided to do layoffs. May he know no peace, wherever he goes.
...but many people in positions where they can start a war or cause some other highly visible event of any sort probably will start turning to Polymarket to make money in the course of their work
Not really, for the same reason entrapment isn't usually seen as an accurate way to gather information for law enforcement. See also Goodhart's law and overfitting.
As a technology company scales up, making great software becomes one of a hundred things the company needs to do right in order to survive and grow. Doesn’t mean it isn’t absolutely essential, but so is having a strong GTM machine, finance competency, operational rigor, HR, and a long list of other essential functions.
It’s only the tech industry where the voice and ego of small companies hold outsized share of voice and love to claim the contrary.
> It’s only the tech industry where the voice and ego of small companies hold outsized share of voice and love to claim the contrary.
I'd be a little bit careful with this claim:
The fact that small companies can have such opinionated opinions without going bust is to me a sign that in particular for software development (but I don't claim that this is transferable to other industries) small teams/companies do have an efficiency advantage.
Many hypotheses can be formulated why this might be the case, like
- software industry is less regulated
- writing good software as the company's product requires a lot less collaboration between many stakeholders than what is necessary for producing other types of sellable products
- in software, "having a smart, though opinionated idea" is of a much bigger advantage (also for the company) than in other, more established industries
Tangential, but companies have been routinely writing and selling software since at least the 1980s, and longer depending on how you draw the line. That's roughly half a century.
At what point will being "less established" stop being an explanation for the way the software industry works?
There's so much content and so many opinions on "meetings" that are polarized and come from a very narrow viewpoint. This particular POV seems reasonably critical of meetings but also recognizing that they are often warranted and can be carried out effectively.
I was happy to see "rallying the troops" as an acceptable reason to have a big meeting, because in my opinion it can be a good reason for a big meeting to exist (even though I'm sure many people disagree.)
I’m ok with age being used as a partial proxy for experience when we’re talking about highly specialized roles with massive implications like the ones that DOGE staffers were dropped into.
> $191k USAGM broadcasting contract for “broadcast operations and maintenance in Ethiopia, Africa”
USAGM's mission is to promote the USA's diplomatic interests in parts of the world with little or no press freedom. Whole thing was cut by executive order of Trump to the maximum extent possible.
Because of that order, it's not even a "not specialised" role, it's not a role.
If USAGM should be cut or not, should have been the choice of congress rather than the executive, but that's a different question entirely.
> Botswana MI curriculum
What's "MI"? Mission-Influenced? That sounds like a plausible amount to spend on a curriculum about Botswana for the benefit of the State Department, let alone in Botswana on anything.
And if it is in Botswana, you have to then actually ask "what is this mission, and is this in the interests of the USA taxpayer?", which needs specialists.
> And if it is in Botswana, you have to then actually ask "what is this mission, and is this in the interests of the USA taxpayer?", which needs specialists.
Specialists in what? Asking whether something is in the interests of the taxpayer? Data analysis? If so, then such specialists would have to be found in an independent organization without conflicts of interest for any specific mission, aligned with the interests of the taxpayer, and they would need to be recruited from the part of the political spectrum that cares about waste in government. In other words, you'd need a group that looks like DOGE.
> Asking whether something is in the interests of the taxpayer?
Yes.
Because they need to:
(1) understand the answer, and not mistake terms of art for things they sound like to normal people. For example, to use Musk's ideology, this would be things like mistaking a study in "transgenic rats" or "trans fatty acids" to be anything about gender.
In the case of `$1.3M State Dept. education contract for “Botswana MI curriculum”`, you've still not said if you recon this is in or about Botswana, and you've still not said what "MI" is, you've taken something that you think "obviously" sounds bad (or why else would you have quoted it?) without having thought too hard. I tried searching, the sidebar was an AI summary of (and linking to) this thread that made claims not supported by anything anyone here has actually said, and only one of the four(!) real links even got me a page with the string "Botswana MI curriculum" on them, which linked to X.com which also didn't explain what that was.
What you've done here is treat it as an applause light, not considered anything about taxpayer interests. Applause lights can be done by an AI, taxpayer interests cannot.
(2) for all items including those that sound good when you do know what they mean, be able to tell if they actually did what they said they did rather than pocket the money.
(3) even when they did the thing, determine if they're any good at doing the thing or if they're a bunch of well-meaning idiots.
For (2) and (3), I'm mainly thinking of the UK with this, with PPE bought for the pandemic that wasn't fit for purpose.
(4) have security clearance to know about clandestine missions, so that you don't cut the expenses which are deliberately faked by the government on purpose with a bland an/or politically correct title so nobody complains about the clandestine mission, despite the money being spend on absolutely nothing at all like whatever the line-item says it was, once what is and isn't "politically correct" gets inverted.
> In other words, you'd need a group that looks like DOGE.
No, you'd get something a lot more competent. And boring.
If you look at those titles and assume that they could be cut, without any more information, you are not a serious person and do not deserve to have any budgetary authority anywhere.
At least bother to come up with some reason they should be cut. But you can't even seem to put that into words.
Apparently "not a serious person" is the new insult of choice with you guys, huh. What a ridiculous reply.
Of course they should be cut. The slogan of the winning party for the last decade was America First. They ran on that platform. Broadcasting and teaching on a different continent isn't putting America First. There's your reason.
The insistence on not understanding obvious stuff is such a tiresome attribute.
> Broadcasting and teaching on a different continent isn't putting America First. There's your reason.
You think advertising doesn't work?
$191k/year to promote American interests in Ethiopia may or may not be value for money to the American taxpayer (I honestly don't care because I'm not one), but to think it can't be value for money is to claim that the primary business model of half the American tech giants — and also the business model of X.com, which isn't a giant but is the property of DOGE's most famous figure-head — is fake.
The US diplomatic agencies, which include USAGM which ran this station, have the business of promoting American interests across the world.
It sells (advertises) the USA's preferences to Ethiopia. Preferences such as "do not interfere with shipping things up the Red Sea or we'll do to you what we did to the Houthis in Yemen". Or preferences like "open your markets to what our businesses want to sell to you". Or, historically, "human rights are in everyone's best interest, you should do more of that because it will make you rich and then you can afford more of our stuff".
Stuff like that.
But to repeat: As I neither know nor care about the national interests of the USA in Ethiopia, I do not say this should or should not be funded — all I say is that this kind of thing *must be considered when deciding if it is or isn't good value, you cannot possibly know a priori just from the title alone*.
The willingness to think you understand and can have an informed opinion on something neither you nor I nor a twentysomething engineer from Tesla know anything about is just as tiresome.
I’m only arguing that there are complex reasons why some of these programs exist and it requires experience and perspective to uncover that and make informed decisions.
