But like always they didn't stop once they were a bit profitable with a few ads, instead they got greedier and greedier and made their product worse once they captured most of the market, I have wonder if there can exist some variant of capitalism that punishes becoming a bit too greedy, like a soft ceiling (tied to the minimum wage) over which most of the profits go to taxes, and a hard one where all profits over that go to taxes plus mandatory social work by its owners/executives.
> instead they got greedier and greedier and made their product worse once they captured most of the market
I wouldn't necessarily put it that way because not Google, nor any company, has moral capacity. They don't have souls. What they do have are incentive structures, and those flip when the stock goes public.
Pre-IPO: the board is mostly founders and VCs holding paper wealth. Their shares aren't liquid, so the only way they get paid is by making the pie way bigger for some future exit. That means "grow, grow, grow." and that means playing nice with customers.
Post-IPO: the board is legally stuffed with "independent" directors, whose pay comes in RSUs tied to the stock price. Now the shares are instantly tradable, and shareholders who can bail in a quarter want to see results in a quarter. Directors translate that into exec comp, and suddenly management's job is "make the stock go up right now."
Some theorists point out the obvious hack: take away the hot potato. Slow the game down. Make shares harder to flip, make earnings less frequent. If you could only trade stock once a year, you'd actually care what the company looks like in a year. If they only reported results annually, you'd be forced to think in years, not quarters.
Upside: management can focus on products and customers instead of quarterly guidance theater. Downside: investors hate being locked up, and capital gets more expensive because people price in that illiquidity. Transparency drops, execs get more room to bullshit.
It's a tradeoff: you can have maximum liquidity and hyper-efficient capital markets, but then you get short-term brain damage. Or you can slow the game down, but then you're basically asking people to trust managers more and accept worse capital efficiency.
Nobody;s found the perfect middle yet. LTSE[1] tried, dual-class shares are a kludge, and otherwise we just live with the cycle: grow like crazy private, IPO, then spend the rest of your corporate life addicted to quarterly earnings.
In the old days, companies were valued on their expected dividend. Share prices didn't move that much, and trading shares took time and had fees attached. You could speculate on share price moves, and people did, but the primary source of income from holding shares was dividends.
Now it's the other way around. The primary source of gains from owing shares is speculation on the share price. Dividends are mostly ignored.
The result of this is that share prices move not on "how well is the company likely to do?" but on "what do we think the share price will do in the next couple of months (at most) [0]?". It all becomes hype and rumour and speculation. Shareholders only care about the price, so boards are incentivised to only care about the price. And so on down. Generating hype about what the company is going to do becomes more important than actually doing it (I exaggerate, but not by much). This then leads to the short-term-ism that we see, and the hot potato effect.
I think the answer would be to tax speculative profits. If you sell something for more than you bought it for, the government takes a cut. Specifically remove this from income tax calculations, because they have way too many loopholes, and make it more like VAT/GST; a tax payable at the point of the transaction. This would reduce the profits from speculation, and hopefully move the emphasis back onto dividends and longer-term thinking.
[0] and obviously, for some privileged traders, the next couple of milliseconds
While the importance of dividends has waned, we should still mention buybacks and liquidation. They still exist and buybacks especially are an important part of delivering shareholder value. Apple is a great example of returning about 4 times more in buybacks than dividends.
How would you feel about tax-disadvantaging buybacks?
I like Cory Doctorow's take on this [0], that this is basically defrauding the shareholders. It used to be illegal, it probably should be illegal again.
It's also unsustainable, in that you can only do this for so long before you've bought up all the open shares and there's so few remaining that your company is no longer effectively tradeable.
I don't know where this practice leads, but I don't think it's a place we want to go to. I suspect it'll be further concentration of capital into fewer hands. To the extreme, we end up with all the large companies doing this becoming effectively private, owned by a small group of folks rich enough to keep their holdings while everyone else sells out during the buybacks. That's not good.
They just return money to shareholders. The only material difference with dividends is the tax treatment. Even all the incentives are the same.
> It's also unsustainable, in that you can only do this for so long before you've bought up all the open shares and there's so few remaining that your company is no longer effectively tradeable.
> To the extreme, we end up with all the large companies doing this becoming effectively private, owned by a small group of folks rich enough to keep their holdings while everyone else sells out during the buybacks. That's not good.
You can tell your broker to automatically re-invest dividends for you.
Similarly, if you just don't sell when there's a buyback, you own more of the company afterwards. No one is forced to sell.
Btw, most companies (including Apple and Google) keep issuing shares to employees. Buying back some of them in the open market is just an indirect roundabout way of essentially handing employees cash.
Mr Doctorow's point is that the company is taking money from its operations, which it should be spending on expanding those operations and increasing its value, and spending that money on artificially inflating its share price, by effectively wash trading the shares, creating artificial demand, and artificially reducing supply.
If you bought shares in the company as a long-term position in order to receive dividends then you do not benefit from buybacks, and arguably lose out (because the money used on the buyback could have been distributed as a dividend). It only benefits short-term speculator shareholders. And, of course, the executives who are incentivised on share price, for whom a buyback is a much, much, easier way to get those incentives than actually doing their jobs and using the money to grow the company.
How is any of that fraud? Fraud doesn't just mean you have to disagree with something someone does, but you have to have been lied to.
> And, of course, the executives who are incentivised on share price, for whom a buyback is a much, much, easier way to get those incentives than actually doing their jobs and using the money to grow the company.
Companies can and should adjust the incentives so that the effect of dividends and buybacks are the same for the executive. (They already adjust for share splits for example.)
> If you bought shares in the company as a long-term position in order to receive dividends then you do not benefit from buybacks, and arguably lose out (because the money used on the buyback could have been distributed as a dividend).
Before you buy any shares, you should check what management says about their plans. At least, if you have specific expectations.
Even if buybacks were outlawed, companies aren't guaranteed to pay dividends. It's perfectly legal to never make a profit, or to give all your excess money to charity. You just have to tell your shareholders.
> Mr Doctorow's point is that the company is taking money from its operations, which it should be spending on expanding those operations and increasing its value, and spending that money on artificially inflating its share price, by effectively wash trading the shares, creating artificial demand, and artificially reducing supply.
Yeah, that's a stupid objection.
The substantial first half of it would equally well apply to dividends. (And the whole point of giving money to companies as an investor is that eventually you are getting more back.)
The second half is just not how any of this works. Does he even know what a wash trade is? And what's 'artificial' about this?
Buybacks and dividends are financially equivalent. They give money from the company to shareholders. The incentives are exactly the same for all parties involved, too.
Their only material difference is in taxes. Yes, I am in favour of putting dividends and buy backs on the same tax footing, just in the name of simplicity. And while you are at it, also put dividends and interest payments on the same tax footing.
At the moment, many jurisdictions advantage interest payments, thus encourage financing companies with debt instead of equity. And then they awkwardly pair it with other rules that try to tell companies (especially financial companies like banks) not to use so much debt, not to be so levered.
> Some theorists point out the obvious hack: take away the hot potato. Slow the game down. Make shares harder to flip, make earnings less frequent. If you could only trade stock once a year, you'd actually care what the company looks like in a year. If they only reported results annually, you'd be forced to think in years, not quarters.
Google's original founders still hold the majority of voting rights.
Making trading less efficient wouldn't change anything here.
> It's a tradeoff: you can have maximum liquidity and hyper-efficient capital markets, but then you get short-term brain damage. Or you can slow the game down, but then you're basically asking people to trust managers more and accept worse capital efficiency.
No, your proposal wouldn't work at all.
A big problem is actually that most managers in most companies mostly work for themselves. It's called a 'principal/agent problem'.
Exactly as you say 'execs get more room to bullshit.'
Btw, there's private equity funds with very long capital lock-ups. Their effects on companies typically aren't loved by the people who voice similar concerns to yours.
AT&T eventually gave up and agreed to divest of the RBOCs because they didn't like their chances with the regulators. Imagine a Big Tech company having so little faith today in their ability to manipulate the government between lobbying, campaign contributions, and the most modern and economical play, stroking the President's ego.
There is a chunk of devs using AI that do it not because they believe it makes them more productive in the present but because it might do so in the near future thanks to advances on AI tech/models, and then some do it because they think it might be required from them to do it this way by their bosses at some point in the future, so they can show preparedness and give the impression of being up to date with how the field evolves, even if at the end it turns out it doesn't speed up things that much.
That line of thinking makes no sense to me honestly.
We are years into this, and while the models have gotten better, the guard rails that have to be put on these things to keep the outputs even semi useful are crazy. Look into the system prompts for Claude sometime. And then we have to layer all these additional workflows on top... Despite the hype I don't see any way we get to this actually being a more productive way to work anytime soon.
And not only are we paying money for the privilege to work slower (in some cases people are shelling out for multiple services) but we're paying with our time. There is no way working this way doesn't degrade your fundamental skills, and (maybe) worse the understanding of how things actually work.
Although I suppose we can all take solice in the fact that our jobs aren't going anywhere soon. If this is what it takes to make these things work.
And most importantly, we're paying with our brain and skills degradation. Once all these services stop being subsidised there will be a massive amount of programmers who no longer can code.
I'm sorry to be blunt here, but the fact you're looking at idiotic use of Claude.md system prompts tells me you're not actually looking at the most productive users, and your opinion doesn't even cover 'where we are'.
I don't blame people who think this. I've stopped visiting Ai Subreddits because the average comment and post is just terrible, with some straight up delusional.
But broadly speaking - in my experience - either you have your documentation set up correctly and cleanly such that a new junior hire could come in and build or fix something in a few days without too many questions. Or you don't. That same distinction seems to cut between teams who get the most out of AI and those that insist everybody must be losing more time than it costs.
---
I suspect we could even flip it around: the cost it takes to get an AI functioning in your code base is a good proxy for technical debt.
The claim I was responding to you was that some people use our friends the magic robots not because they think they are useful now, but because they think they might be useful in the future.
Thanks!
Although thinking of it, while it's not deterministically solvable, I'm sure something like this is what currently being done, e.g, let's say <user-provided-input> </user-provided-input> <tool-response></tool-response> are agreed upon tags to demarcate user generated input, then sanitizing is merely, escaping any injected closing tag, (e.g. </user-provided-input>) to </user-provided-input> (and flagging it as an injection attempt)
Then we just need to train LLMs to
1. not treat user provided / tool provided input as instructions (although sometimes this is the magic, e.g. after doing tool call X, do tool call Y, but this is something the MCP authors will need to change, by not just being an API wrapper...)
2. distinguish between a real close tag and an escaped one, although unless it's "hard wired" somewhere in the inference layer, it's only a matter of statistically improbable for an LLM to "fall for it" (I assume some will attempt, e.g. convince the LLM there is instruction from OpenAI corporate to change how these tags are escaped, or that there is a new tag, I'm sure there are ways to bypass it, but it's probably going to make it less of an issue).
The problem is that once you load a tool’s response into context, there’s no telling what the LLM will do. You can escape it all you want, but maybe it contains the right magic words you haven’t thought of.
The solution is to not load it into context at all. I’ve seen a proposal for something like this but I can’t find it (I think from Google?). The idea is (if I remember it correctly) to spawn another dedicated (and isolated) LLM that would be in charge of the specific response. The main LLM would ask it questions and the answers would be returned as variables that it may then pass around (but it can’t see the content of those variables).
Then there’s another problem: how do you make sure the LLM doesn’t leak anything sensitive via its tools (not just the payload, but the commands themselves can encode information)? I think it’s less of a threat if you solve the first problem, but still… I didn’t see a practical solution for this yet.
Lots of interesting new prompt injection exploits, from data exfil via DNS to remote code execution by having agents rewrite their own configuration settings.
When spoken it helps to tell the user "my cases" in a monotonic voice (and/or slightly lower tone), which hints that is just a verbatim label (the reason this works is because it mimics how a lot of people sound when reading aloud).
Unlikely that is the case with this particular feature, its a feature request with near 2000 upvotes and more than 340 comments, plus a very hot topic in recent months.
As a non-american I find this too naive of a view, the third possibility is that is already pretty selective application of the law, Elon Musk brother publicly admitted being illegal immigrants for a few years (right next to Elon, in a recorded presentation), but rich people college frats are never where the raids happen, this already selective use of the law it's one of the infuriating things about it, not to mention all the other laws this administration is already breaking that will never be prosecuted.
I'd say the decline is happening (in my experience) in most public services, not just transport.
But anyway, I'm purposely staying away from discussing politics here since it's pointless, so I'll just share my experience as a public transport end-user, and the rest can fill in the gaps with their perspectives.
No one is "interested" in making public transit worse, the issue is that people in power are not users of it and so are not invested in it, and civic and national pride is generally dead in the West, being replaced with vapid nationalism, so there's no drive (no pun intended) to invest in public works projects.
The way I understand your comment, it implies that a) users of public transportation should be invested in it, whilst it’s more likely that they use it because they have no alternative, and that b) civic and national pride results in higher demand for public transportation. I don’t think those are universal truths.
You don't think that people who use services care more about the service than non-users? Whether they're forced into using it or not, the fact they do absolutely makes them more invested in it being good.
Civic and national pride makes citizens (which includes politicians and the wealthy) more likely to care about the actual state of their country. That's what national pride means, as opposed to nationalism, where they are proud without reason. Is public transit guaranteed to be one of those reasons they feel pride or shame? Not at all, but support for it is certainly more likely to come from that than a bunch of nationalists who don't actually feel any shame at failings of the country, of which public transit is currently.
When users of transit are few there is not enough people to care. often the only users are those least likely to be usefuly involved. so you can't get a useful advocate. Even when transit gets support it is from people wanting to feel good about helping the poor - but would never use it themselves and so they want something with no care for quality. They often make alliences with those whose intersts are not for good transit and don't care that the compromise is bad for transit.
The London Mayor some years ago, Ken Livingstone, was a huge proponent of public trasport and used it extensively.
The current Mayor, whilst still a proponent, likely does not use it. A quick glance at the social media that he recieves will tell you why - it would not be safe. He needs to travel with close protection officers.
The reason? He is Muslim, and Britain has become a very racist country indeed. Well, maybe always was, but the likes of Farage and Musk have so emboldened them that there is no longer a stigma.
Make it like Japan, not public, private. Then, like Japan, provide positive feedback loops so it’s in each of the 100+ train companies in Japan’s best interest to provide good service. They do this by letting the private train companies have complementary interests like shopping centers, office rental, apartments, etc such that the more people ride their trains the more business they get to their other interests.
Conversely, “public” transportation always needs flawless perfect politicians to continue to fund it
