That was obvious before even looking at the repo because the OP used "the core insight" in the intro. Other telltale signs of these type of AI projects:
- new account
- spamming the project to HN, reddit etc the moment the demo half works
- single contributor repo
- Huge commits minutes apart
- repo is less than a week old (sometimes literally hours)
- half the commits start with "Enhance"
- flashly demo that hides issues immediately obvious to experts in the field
- author has slop AI project(s)
OP uses more than one branch so he's more sophisticated than most.
> I suspect big sellers must have dedicated account managers
No clue how it is elsewhere but in Amazon India, the largest seller is Amazon itself which sells under a different name. That's their model. They were under scrutiny by the Indian Government [1] [2] [3] last time I checked. Keeps registering subsidiaries under different names.
So you are basically competing with Amazon itself, which also acts as a seller in their own store.
India will be chairing BRICS this year. And Trump decided to fuck with the wrong country, especially when India-US relations were at its peak. This year will mark the beginning of the end of USD as a global reserve.
There is hardly anything that is made domestically in the US. So the premise falls apart almost immediately. This premise works great for India where domestic production exceeds exports by massive margins and the economy depends mostly on domestic economy. It does not work for US where there is hardly any domestic production and is totally import driven economy.
The US has the second largest manufacturing base in the world after China. It's larger than India and is even slightly larger than the EU. It used to be the largest. Moreover, if the premise is that you're trying to bring back manufacturing capacity, it doesn't matter if something is currently made in the US, what matters is what it would cost if it was, because a tariff in excess of the difference would then make that cost effective.
Obviously in the latter case you would then have to wait until that manufacturing capacity comes back online, but "customers switch to a domestic product" isn't the only thing that can cause foreign manufacturers to have to lower prices. They could also switch to substitute products or reduce consumption and then foreign manufacturers would still have to lower prices to limit the extent to which that happens.
> The US has the second largest manufacturing base in the world after China. It's larger than India and is even slightly larger than the EU.
Of only end products. The "largest manufacturing base" is misleading when majority of inputs for your finished goods are dependent on imports.
> it doesn't matter if something is currently made in the US
It definitely does matter. The right way to have gone about this was to first build the manufacturing capacity in US before imposition of tariffs. It was done in the reverse, which led to US revealing its hand too early, allowing for rest of the World to re-calibrate and start the process of de-dollarization.
> foreign manufacturers to have to lower prices. They could also switch to substitute products or reduce consumption and then foreign manufacturers would still have to lower prices to limit the extent to which that happens.
It won't happen. There is no reason for exporters to lower prices when tariffs only set a new normal. Once the prices have gone up and consumer spending has stabilized around those jacked up prices, that will set the benchmark. Just study history. No product has been devalued due to any contingent circumstances unless the product itself becomes obsolete. Here you are not talking about novel products being developed and manufactured that will obsolete something popular. You are talking about bringing back manufacturing of nuts, bolts etc. Things that are critical and have an already established price in the market that will only go up higher in price once manufacturing moves to US eventually. Rest of the World will adjust to the new higher price.
> Of only end products. The "largest manufacturing base" is misleading when majority of inputs for your finished goods are dependent on imports.
Isn't it easier to figure out how to manufacture screws given steel than to figure out how to manufacture airplanes given screws?
> The right way to have gone about this was to first build the manufacturing capacity in US before imposition of tariffs.
But what's the incentive to do that if there are no tariffs? You either need a reward for doing it (subsidies) or a penalty for not doing it (tariffs). But the US government is already running unsustainable deficits, so there is no money for subsidies unless you want to pay higher taxes or cut some other spending, and good luck convincing people to do either of those things.
> There is no reason for exporters to lower prices when tariffs only set a new normal.
If China charges $100 to manufacture something that the US would have to charge $120 to manufacture and then you put a >$20 tariff on it, China could keep charging the same prices, but that would make it cost effective to make it in the US. To prevent that from happening they would have to lower the price, and therefore have the incentive to.
Now suppose it would cost $200 to manufacture in the US and the tariff is 50%. Can China just keep the price where it is and make the customer pay $150? The customer has a finite amount of money, so if they did then the customer would buy a new phone every six years instead of every four years. Meanwhile the $1000 phone just became cost effective to manufacture domestically, because 50% of that is more than the $100 increase in manufacturing cost, so they lose that business too. What response to this do they have other than to lower prices?
China has a third of global manufacturing capacity (and have to run flat out to avoid deflation due to domestic consumption that will never grow to meet domestic production capacity). Only the unsophisticated could believe the US is going to increase domestic manufacturing capacity at the levels needed to make domestic sourcing superior in some manner. That’s why these tariffs are not grounded in reality.
In three years at the most, these tariffs are done. Cheaper to eat the premium in the short term versus suboptimally invest capital in long duration investments (ie local factories and equipment to fill them). Manufacturing jobs continue to decline, as they have since the election.
> China has a third of global manufacturing capacity (and have to run flat out to avoid deflation due to domestic consumption that will never grow to meet domestic production capacity).
This sounds like the case that they'd end up paying more of the tariffs, because they can neither reduce production nor absorb the output domestically, which leaves lowering prices to try to sustain the same volume.
> In three years at the most, these tariffs are done.
That's not happened the last time. Trump was out for four years and Biden kept a lot of the tariffs.
> Manufacturing jobs continue to decline, as they have since the election.
"Manufacturing jobs" are a talking point but were never really going to happen. The only way the US is actually going to compete is by increasing automation, which doesn't do much for "manufacturing jobs". But it does get the manufacturing into the same timezone as the people designing the products to prevent those jobs from going next, keep the knowhow of production active domestically, reduce dependence on China, etc.
This is a graph showing that construction spending on manufacturing in the US is more than double what it was a decade ago and trending only slightly down from the all-time high in the period trailing when the COVID money ran out and the Fed started to raise interest rates.
While in the abstract and academic sense this is true, in practice there are two big problems that make it an utter non-starter*:
1) Due to the absolutely massive supply chains that have been built up in East Asia (not just China, but many other countries around there), and lack of same in the US, even for products where it's physically possible to produce it all domestically, from the raw materials on up, it would take decades of sustained investment without return before actual consumer products could be made on anything other than a one-off basis. Any step that can't be done in-country gets the tariffs slapped on again. And there are a fair number of raw materials we just don't have, at least not in the kinds of amounts that, um, the entire rest of the world does, that are required for mass production.
2) Trump isn't applying tariffs in a strategic manner to get domestic manufacturing to come back. He's applying tariffs as his personal punishment stick, and to all appearances that's the best he's actually capable of doing with them. In order for any of what I described in #1 to happen, ever, the tariffs need to be applied consistently, predictably, and for a long time.
Trump doesn't want to do any of that. He's just found a magic stick that makes people kowtow to him, and he's going to use it however he pleases.
* Not that I think you're unaware of these, based on your post; to a large extent I'm just expanding upon your second paragraph here.
> it would take decades of sustained investment without return before actual consumer products could be made on anything other than a one-off basis.
That's true of some products, not all of them, or even a majority. And even for those products, well, if it's going to take a long time then we better get started.
> And there are a fair number of raw materials we just don't have
This is again not the common case, and even then it's not necessarily the wrong solution. For example, China currently dominates the production of rare earths and the US doesn't have sufficient reserves, but Australia does, so higher tariffs on China than Australia create an incentive to move mining operations to Australia which breaks China's lock, and creates the incentive to invest in rare earth processing in the US, since then you're only paying the (lower) tariff on the (lower-priced) raw materials rather than the (higher-priced) refined product.
> Trump isn't applying tariffs in a strategic manner to get domestic manufacturing to come back.
This is more of a Trump problem than a tariff problem. If you do something wrongly enough it obviously doesn't work as well as it otherwise might.
> China currently dominates the production of rare earths and the US doesn't have sufficient reserves, but Australia does, so higher tariffs on China than Australia create an incentive to move mining operations to Australia which breaks China's lock
There are "sufficient reserves" (known rare earths in the ground) across the globe and the US absolutely has large reserves.
> to move mining operations to Australia which breaks China's lock
There are already mining operations in Australia delivering raw concentrates in bulk to China. Again, not a shortage of mining operations or a shortage of reserves in the ground.
It's the concentrate processing that China invested time and capital in decades past - every other country about the globe (save for Malaysia, to their regret) figured they'd leave the acres of acid ponds and low level radioactive waste to the Chinese.
Now the US wants Australia to take that on, and that's a deal with the devil for Oz while the current POTUS cannot be trusted to hold up any deal.
Far from getting a break, you guys are paying tax on tax. You indirectly pay for import taxes every time your companies import raw materials needed to finish their goods (added value) and then that final value (cost of import + added value) has its own sales tax. AFAIK there are no input credits for US sales tax. Then you also have VAT but at least VAT is only on the added value.
Income tax is way better as you can reduce the tax burden by including expenses/deductions. You cannot do the same for tariffs, sales tax and VAT as an end consumer. VAT is only beneficial to businesses as they can subtract inputs from outputs.
> You indirectly pay for import taxes every time your companies import raw materials needed to finish their goods (added value) and then that final value (cost of import + added value) has its own sales tax.
This isn't really any different than any other kind of taxes. You pay income tax and then pay sales tax using the money that was already taxed as income. The construction company pays sales tax when it buys a backhoe, which increases construction costs and therefore real estate prices, and then you pay property tax on the higher real estate prices, and make the bigger mortgage payment with money that was already taxed as income.
The only way you'd really get something different with tariffs is if the supply chain for some product passes through the local country multiple times, i.e. it gets imported, exported and then imported again. Which probably happens occasionally but isn't the common case.
Meanwhile how many times something is taxed isn't really the relevant thing. It's, how much in total are you paying in taxes? If you pay ~10% three times, that's not really any worse than paying ~33% once. It is, of course, worse than paying 10% once.
> This isn't really any different than any other kind of taxes. You pay income tax and then pay sales tax using the money that was already taxed as income
It definitely does make a huge difference. From sibling comment I got to know US does not even have VAT. It only makes the situation worse as the businesses operating in US cannot offset input credits against their output liability as Sales Tax has no such concept. So you are paying tax-on-tax-on-tax all the way to your raw materials that have been imported APART from paying tariffs. No wonder prices are so jacked up in US and to compensate that, you all have inflated salaries. The US Government is fleecing its citizens dry. Please study how VAT/GST works in EU/India/Australia and compare it with Sales Tax regime in US and you will know why Sales Tax is so bad.
> Meanwhile how many times something is taxed isn't really the relevant thing. It's, how much in total are you paying in taxes? If you pay ~10% three times, that's not really any worse than paying ~33% once. It is, of course, worse than paying 10% once.
You are not paying 10% three times. Assuming raw material was imported at $X + %10 of $X (tariff is 10%), value add was say $10, then the IRS is collecting say sales tax of 10% of the total value: 10 % of (($X + %10 of $X) + $10). Now this is just the simplest chain where raw material -> imported by manufacturer -> sold directly to consumer. But that is not how it is done. You typically buy from a retailer who buys from a dealer who buys from a wholesaler/manufacturer. So that would be 10% every time ON THE FULL VALUE (not just on value added).
To demonstrate a simple raw material -> imported by manufacturer with value added -> sold to dealer/distributor -> sold to retailer -> sold to customer, this is what it would look like:
1. Imported by manufacturer:
$X + %10 of $X
2. Value added ($10) and sold to dealer/distributor:
10 % of (($X + %10 of $X) + $10)
3. Dealer stocking/shipping charges added (say $10 again) and sold to retailer:
10% of (10 % of (($X + %10 of $X) + $10) + $10).
4. Retailer stocking/service charges added (say $10 again) and sold to consumer:
10% of (10% of (10 % of (($X + %10 of $X) + $10) + $10) + $10).
The longer the chain, the more tax-on-tax you are paying (in some cases the total final tax can even go above the actual cost of making the product). This nonsense is solved by VAT/GST where the tax you pay for acquiring raw material or processed inputs comes back to you as input credits, which you can use to offset your output tax liability. There is no compounding of tax in VAT/GST.
My answer is based on typical definition of what sales tax is. I have no idea how it is actually implemented in US and how it avoids tax pyramiding and if it actually can or not. But I know for a fact that ST has no concept of input tax credits unlike VAT/GST.
EDIT: Turns out US sales tax indeed works like I described above [1]. There are definitely some instances where B2B sales can be exempt from paying sales tax but it does not seem to be pervasive enough to be worth highlighting: because supposedly 40% of business transactions are taxed at intermediate stages (are paying sales tax at every stage).
Quoting from article:
"Some studies estimate that around 40% of the total sales tax revenue comes from taxes levied on business-to-business sales."
> because supposedly 40% of business transactions are taxed at intermediate stages (are paying sales tax at every stage)
That's not what that says.
If a business buys office furniture because it wants to furnish their offices rather than because they're in the business of selling office furniture then they pay sales tax on it even though it's a "business-to-business" transaction. This isn't any different than when they buy real estate so they have offices and then pay property tax on it. They're the end customer and the end customer pays the tax.
This results in "double taxation" (they pay the tax on the furniture, they pass the cost on to their customers as higher prices, the customers pay more tax on the higher prices), but that's the same as any other overlapping set of taxes.
What it doesn't do is cause longer supply chains to pay more in taxes, because that's the case where they can get the exemption when they're buying something for resale.
You really don't need ANY study to know the obvious thing that tariffs are taxes on imports. It is being paid by Americans from DAY 1. The only difference being American Companies were taking the brunt of it all and it is now obvious that they cannot keep swallowing it and will eventually pass it on to the American consumers.
It is crazy that so many in US STILL think tariffs are being paid for by exporting countries.
US is sabotaging itself and pushing in the same "New World Order" that the right-wing conspiracy nuts kept warning about but ironically have been instrumental in accelerating it themselves anyways.
Or maybe that was the design all along. To not go out in a whimper but with a big bang.
If I were the Democrats, I would do nothing and just let the US admin destroy whatever little credibility it has left on the World stage... thereby securing mid-terms and the next Presidential elections.
> If I were the Democrats, I would do nothing and just let the US admin destroy whatever little credibility it has left on the World stage... thereby securing mid-terms and the next Presidential elections.
Not a great idea. The US will have lost a lot of political goodwill by then. And given up a lot of geopolitical status and influence. The devices will be back in the saddle but have to resort to really unpopular measures to clean up the mess, basically guaranteeing a republican win afterwards.
And some credibility and influence will never recover. The rest of the world will remember there can always be another trump. And they will have switched to (and restarted) local industries. Once those are running there's no incentive to look at US ones again. And any geopolitical influence that was lost will already have been filled by other players who will entrench themselves.
> The US will have lost a lot of political goodwill by then
Not really. All Governments across the World have undergone similar crisis in their own Countries, in different points in time in their own history, and know well to differentiate an erratic Government from its populace. I bet no one has ill-will towards Americans (unless you are into terrorism) and so will obviously want to mend ways once a better admin is elected and put in place.
The issue really though is not about political goodwill. It is about business that gets moved elsewhere. Once moved, it is really hard to bring it back. So US will then have to make a lot of concessions when it comes to that. The reason rest of the World is upset with the new tariff regime is not because they are being affected by it monetarily (which they actually are not), but because they do not want to sever a well established system and go looking for new buyers/sellers. For example, China decided to stop buying soybeans from US (after Trump threatened tariffs) and switched to buying from Argentina. Now once a new US admin is in place, it will have to give a better deal for China to switch its already established supply chain with Argentina to US again. Inertia is a good thing in trade. Infact, I feel it is even more important than building a moat. People are willing to continue paying higher price (even if something cheap is available elsewhere) because they do not want to go through the hassle of moving away from already established supply-chains and renegotiate new deals.
> really unpopular measures to clean up the mess
Agreed.
> Once those are running there's no incentive to look at US ones again
They will if US comes with better deals. This is the "unpopular measures to clean up the mess" you were pointing to earlier. Right now there is a mismatch between what US is actually worth and what it is projecting. I am not saying it is not a superpower anymore. It still very much is. But it is failing to recognize the rise of great powers in the rest of the World. And is unable to reconcile with multi-polarity that has already arrived and exists. As long as US is in denial of this reality, it will continue to make mistakes.
> And any geopolitical influence that was lost will already have been filled by other players who will entrench themselves
I agree but I'll go further and argue that the Trump admin is behaving erratically because it has realized it has lost geopolitical influence and that other players have already entrenched themselves. Trump has diagnozed the problem correctly (that US has lost its hegemony in many areas) but has no expertise (or even experts around him) to advice him on the right course of action.
> Not really. All Governments across the World have undergone similar crisis in their own Countries, in different points in time in their own history, and know well to differentiate an erratic Government from its populace. I bet no one has ill-will towards Americans (unless you are into terrorism) and so will obviously want to mend ways once a better admin is elected and put in place.
The governments probably yes but what will be lost for much longer is the heart of the people. In Western europe we really idolised American society. Most movies in the cinema were (and still are) American. A lot of music is. There is a huge American influence on western European society.
However right now that idolisation has turned to irritation and eyerolling. I mean, if America was that great, how could it fall so quickly like that? And if Trump starts a war over Greenland that will turn worse pretty quickly. And those things take much longer to revert. Like they say, trust arrives on foot and leaves on horseback.
I remember when I was in Holland in the 80s, the Germans were still deeply detested for what they had done to us 40 years earlier. And any cultural German phenomenon was tainted by it. And with that buyin to German economic developments and products, political proposals etc. And their tourists "conquering" our beaches with their towels. And yes we did make exceptions for good Germans, just like we know there are good Americans too. But in general the feeling was pretty negative and it took a long time to mend. A government can strike a peace deal pretty quickly but people don't just turn around like that.
In the 90s that dislike for the Germans really started to turn and it's pretty much gone now but it remained for a very long time.
> They will if US comes with better deals. This is the "unpopular measures to clean up the mess" you were pointing to earlier. Right now there is a mismatch between what US is actually worth and what it is projecting. I am not saying it is not a superpower anymore. It still very much is. But it is failing to recognize the rise of great powers in the rest of the World. And is unable to reconcile with multi-polarity that has already arrived and exists. As long as US is in denial of this reality, it will continue to make mistakes.
The US still is a military superpower but it can't maintain that without the hearts of the people. Again look at Germany, they were by far the most powerful military nation in the world. Yet they were defeated because the rest of the world got sick of them and they had to go. The US even joined forces with China and Russia, as unlikely partners as they were even in those days.
The US' hegemony was a carefully constructed combination of military power, economic power (the petrodollar in particular) and cultural adoration. Two of those are already crumbling and it's using the military power to forcefully hold on to its position. That is not sustainable.
Some of the things Trump always complained about like the overwhelming majority of money for NATO coming from America, that was not just laxness from Europe, it was by design. America didn't want a too powerful Europe. They didn't want Europe to have its own nuclear umbrella. It bought influence. But you can't pull too hard on that. This is why you see that influence crumbling so fast.
> It is crazy that so many in US STILL think tariffs are being paid for by exporting countries.
They knew it was a lie then, and they know it's one now. A plurality of voters want what's happening currently, they're not crazy, it's just a mix of xenophobia, isolationism, and inbreeding.
The problem with that last is that if the Democrats do nothing, there's a very real chance that Trump deploys the military to either prevent the midterms from happening, or to force their outcome to be what he wants.
(Of course, if they do something there's still a chance he does that—they have to do the right thing and they have to do it well to reduce the chances by much!)
There's also that pesky matter of, y'know, their constituents. Who are getting bled dry by the stagflation that's happening.
> You really don't need ANY study to know the obvious thing that tariffs are taxes on imports. It is being paid by Americans from DAY 1. The only difference being American Companies were taking the brunt of it all and it is now obvious that they cannot keep swallowing it and will eventually pass it on to the American consumers.
I say they exact same thing to my liberal friends who endorse corporate tax increases, but they don't think it is obvious that companies will eventually pass it on to the American consumers.
Corporate tax or even personal income tax is not the same as tax on cost of the product itself. As a corporate, I can structure my expenses in such a way that I never pay tax irrespective of whether the tax rate is 10% or 50%. There are ways to do that (through deductions/expenses/investments in assets that can be depreciated and included as expenses etc) and even defer my earnings over a period of time. You have various ways to reduce final tax liability.
You cannot escape paying tariffs or sales tax as it is levied on the product you are purchasing. It is not a tax that is levied AFTER you have calculated your income.
I wouldn't say it is a tree as such as at least trees are deterministic where input parameters (seed, environment, sunlight) define the output.
LLM outputs are akin to a mutant tree that can decide to randomly sprout a giant mushroom instead of a branch. And you won't have any idea why despite your input parameters being deterministic.
You haven't done a lot of gardening if you don't know plants get 'randomly' (there's a biological explanation, but with the massive amounts of variables it feels random) attacked by parasites all the time. Go look at pot growing subreddits, they spend an enormous chunk of their time fighting mites.
Determinism is not strictly anti-randomness (though I can see why one can confuse it to be polar opposites). Rather we do not even have true randomness (at least not proven) and should actually be called pseudorandom. Determinism just means that if you have the same input parameters (considering all parameters have been accounted for), you will get the same result. In other words, you can start with a particular random seed (pseudorandom seed to be precise) and always end up with the same end result and that would be considered deterministic.
> You haven't done a lot of gardening if you don't know plants
I grow "herbs".
> there's a biological explanation
Exactly. There is always an explanation for every phenomena that occurs in this observable, physical World. There is a defined cause and effect. Even if it "feels random". That's not how it is with LLMs. Because in between your deterministic input parameters and the output that is generated, there is a black box: the model itself. You have no access to the billions of parameters within the models which means you are not sure you can always reproduce the output. That black box is what causes non-determinism.
EDIT: just wanted to add - "attacked by parasites all the time", is why I said if you have control over the environment. Controlling environment encompasses dealing with parasites as well. Think of well-controlled environment like a lab.
Do you think LLMs sidestep cause and effect somehow ? There's an explanation there too, we just don't know it, But that's the case for many natural phenomena.
In what world are trees deterministic? There are a set of parameters that you can control that give you a higher probability of success, but uncontrollable variables can wipe you out.
Explained here [1]. We live in a pseudorandom World. So everything is deterministic if you have the same set of input parameters. That includes trees as well.
I am not talking about controllable/uncontrollable variables. That has no bearing on whether a process is deterministic in theory or not. If you can theoretically control all variables (even if you practically cannot), you have a deterministic process as you can reproduce the entire path: from input to output. LLMs are currently a black box. You have no access to the billions of parameters within the model, making it non-deterministic. The day we have tools where we can control all the billions of parameters within the model, then we can retrace the exact path taken, thereby making it deterministic.
Humans writing code are also non deterministic. When you vibe code you're basically a product owner / manager. Vibe coding isn't a higher level programming language, it's an abstraction over a software engineer / engineering team.
That's not what determinism means though. A human coding something, irrespective of whether the code is right or wrong, is deterministic. We have a well defined cause and effect pathway. If I write bad code, I will have a bug - deterministic. If I write good code, my code compiles - still deterministic. If the coder is sick, he can't write code - deterministic again. You can determine the cause from the effect.
Every behavior in the physical World has a cause and effect chain.
On the other hand, you cannot determine why a LLM hallucinated. There is no way to retrace the path taken from input parameters to generated output. At least as of now. Maybe it will change in the future where we have tools that can retrace the path taken.
You misunderstand. A coder will write different code for the same problem each time unless they have the solution 100% memorised. And even then a huge number of factors can influence them not being able to remember 100% of the memorised code, or opt for different variations.
People are inherently nondeterministic.
The code they (and AI) writes, once written, executes deterministically.
> A coder will write... or opt for different variations.
Agreed.
> People are inherently nondeterministic.
We are getting into the realm of philosophy here. I, for one, believe in the idea of living organisms having no free will (or limited will to be more precise. but can also go so far as to say "dependent will"). So one can philosophically explain that people are deterministic, via concepts of Karma and rebirth. Of course none of this can be proven. So your argument can be true too.
> The code they (and AI) writes, once written, executes deterministically.
Yes. Execution is deterministic. I am however talking only about determinism in terms of being able to know the entire path: input to output. Not just the outputs characteristic (which is always going to be deterministic). It is the path from input to output that is not deterministic due to presence of a black box - the model.
I mostly agree with you, but I see what afro88 is saying as well.
If you consider a human programmer as a "black box", in the sense that you feed it a set of inputs—the problem that needs to be solved, vague requirements, etc.—and expect a functioning program as output that solves the problem, then that process is similarly nondeterministic as an LLM. Ensuring that the process is reliable in both scenarios boils down to creating detailed specifications, removing ambiguity, and iterating on the product until the acceptance tests pass.
Where I think there is a disconnect is that humans are far more capable at producing reliable software given a fuzzy set of inputs. First of all, they have an understanding of human psychology, and can actually reason about semantics in ways that a pattern matching and token generation tool cannot. And in the best case scenario of experienced programmers, they have an intuitive grasp of the problem domain, and know how to resolve ambiguities in meatspace. LLMs at their current stage can at best approximate these capabilities by integrating with other systems and data sources, so their nondeterminism is a much bigger problem. We can hope that the technology will continue to improve, as it clearly has in the past few years, but that progress is not guaranteed.
Agree with most of what you say. The only reason I say humans are different from LLMs when it comes to being a "black box" is because you can probe humans. For instance, I can ask a human to explain how he/she came to the conclusion and retrace the path taken to come to said conclusion from known inputs. And this can also be correlated with say brainwave imaging by mapping thoughts to neurons being triggered in that portion of the brain. So you can have a fairly accurate understanding of the path taken. I cannot probe the LLM however. At least not with the tools we have today.
> Where I think there is a disconnect is that humans are far more capable at producing reliable software given a fuzzy set of inputs.
Yes true. Another thought that comes to my mind is I feel it might also have to do with us recognizing other humans as not as alien to us as LLMs are. So there is an inherent trust deficit when it comes to LLMs vs when it comes to humans. Inherent trust in human beings, despite being less capable, is what makes the difference. In everything else we inherently want proper determinism and trust is built on that. I am more forgiving if a child computes 2 + 1 = 4, and will find it in me to correct the child. I won't consider it a defect. But if a calculator computes 2 + 1 = 4 even once, I would immediately discard it and never trust it again.
> We can hope that the technology will continue to improve, as it clearly has in the past few years, but that progress is not guaranteed.
Really sucks to see this happen! Been using Tailwind for past few years now.
All the more reason to go closed source. Except for few really vital components that have national security implications (OS/Kernel, drivers, programming languages), which can be funded and supported by universities, Governments etc, I am of the strong opinion that everything else should go closed source.
People with that perspective shouldn't have been doing open source in the first place. AI isn't hurting people sharing things, only people who are pretending to share but actually indirectly selling things.
There is no one in this World who will do things purely for altruistic purposes. Even if not for money, it would be for something intangible that ingratiates the Self (fame for example).
I can't find a single example of a software developer who has put out software purely for some altruistic purpose without any returns on that investment (direct or indirect).
Building a sustainable business model was a great way to justify open source. Not anymore.
> There is no one in this World who will do things purely for altruistic purposes. Even if not for money, it would be for something intangible that ingratiates the Self (fame for example).
And pretty much none of that is threatened by AI. LLMs learning from code, or articles, found online, are at worst neutral to this, and in many ways beneficial. They're only negative if you're using your contribution as bait to hold your audience, in lieu of a more honest approach of openly selling it to them.
> Building a sustainable business model was a great way to justify open source. Not anymore.
Or poison it. Open Source as a business model was always an anti-competitive hack, similar to "free with ads": drive price down to 0, destroying any other way of making money in this space, including "honest exchange of money for value".
[1]: https://github.com/ChartGPU/ChartGPU/blob/main/.cursor/agent...
[2]: https://github.com/ChartGPU/ChartGPU/blob/main/.claude/agent...
reply