Would you buy a product with asbestos in if it wasn't regulated against? (Assuming we removed all similar regulation so the lack of regulation does not its self imply safety.)
Of course you wouldn't.
We have journalists to uncover dangers like this; they are clearly financially incentivised to do so. We have courts to assess damages. We don't need government regulation.
Such a common trope that "the heartless capitalist doesn't care about harming customers so we need the government to save us". Of course the capitalist cares about harming customers, she needs to sell to them (and their competitors product will be much more successful if it is not harmful!).
And, in either case, regulation or free market will only save us if there are viable alternatives. Fossil fuels still kill people, but we don't regulate against it because there is currently no viable alternative.
I don't know if this comment is one of ignorance, or juvenile "well actually", but it is tragically misinformed. From an Australian perspective all of the big players, CSR, John Mansviille, & James Hardy, knew asbestos was a significant hazard by _at least_ the 40s. There were early epidemiological studies of cancers around asbestos work sites, and workers, in the 50s here in NSW. Unions and gov health departments start to push back on exposure and seek meaningful damages in the 60s and 70s. There were _public_ campaigns about the dangers in the 70s and 80s. It wasnt meaningfully restricted, _and continued to be commonly used_, through the 80s. A complete ban, primarily workplaces IIRC, wasnt introduced until 2003. The randian wank fantasy of "the informed consumer knows best" has been repudiated innumerable times.
And, as others have pointed out, this is not an individual choice. The families who got asbestosis from washing their fathers work clothes didnt make a choice. The bloom of cancers for residential suburbs miles around james hardy in camellia didnt have a choice. There is no expiration date on the dangers of friable asbestos. It remains hidden in the common environment forever, until someone else stumbles on it.
Would you buy a product with asbestos in it if its presence wasn't disclosed? You might, if it provided value vs alternatives.
Manufacturers are successuful when they sell. If their product is found dangerous they a) deny and muddy the waters, b) settle lawsuits and if that doesn't work c) close up shop and open a new business. Customer unwelfare is a cost of business.
Asbestos is hard to hide, certainly competitors of the Asbestos-using company would know they were using Asbestos. They, at least, would have an incentive to advertise how bad their competitors product is because it uses Asbestos.
Equally, once it has been established that Asbestos is harmful any company using it would be so sued that they would quickly cease to exist.
Yes the free market doesn't stop health risks immediately but neither does regulation (see: asbestos!)
Both regulation and an actually healthy free market are important.
Both have pros and cons
Regulation sucks at directing productions of goods and setting prices, distributing these tasks to the people generally works better.
The free market is unstable though, its actors ever trying to gain advantage and squash competition. Profit first driven people too often push harm to humans as out of mind and externalized.
This is where regulation is needed. (read as: we the people need rules to protect what we deem important, including our health and having a well-functioning society, profits be damned)
Would you tell people your product has asbestos in it, if it wasn’t regulated that you had to? Of course you wouldn’t. And then, apparently, you would blame your customer for buying it.
No doubt when electric cars become better in every way than fossil fuel cars, the government will create a regulation banning fossil fuel cars. Everyone will rejoice that the government stopped those horrible fossil-fuel burning cars! Of course, the vast majority of people would have switched to the electric cars at this point anyway.
Internal memos from 1970s - 2000s reveal knowledge of asbestos traces. Executives, better known as caring capitalists, resisted disclosure.
In case you think this isn’t a pattern: Purdue and Oxy, General Motors and their ignitions switches, DuPont and “confirmed animal carcinogen” Teflon, Philip Morris’ cigarette campaigns, VWs dieselgate…
Each of them found to be suppressing knowledge knowingly harming consumers.
Yes of course it is in their interest to lie / hide until it becomes public knowledge. But regulation has the exact same problem (hence why Asbestos was not regulated for a long time, same with your other examples).
When it is widely known (and therefore can be regulated) it is already too late as it's now in the producers interest to cease producing it.
Heavy litigation after the fact can disincentivise the lying in the first place. If legislation doesn't not allow for this (e.g. because of time limitations) then it should be amended.
All successful markets in history have been regulated to some degree. A market requires sellers AND buyers and buyers will flock to a market where they're not burdened with the cost and expense of having to do their own research and investigation before even considering a purchase, never mind actually completing a trade.
Completely unregulated markets simply don't scale in terms of successful trading - the regulation replaces the work each buyer would have to do and thus is actually more efficient than having each buyer replicate the work of the regulator. This is why they have been out-competed by regulated markets in the course of history.
I don’t understand your point - there’s no symmetry there.
Regulated venues dominate nearly every sphere of trading in terms of volume.
They have done this by being more attractive to buyers and sellers than unregulated venues - i.e. regulated venues have out-competed unregulated venues.
In the market for trading venues, regulated venues have won.
Complete nonsense. The free market will tell him if he can do this or not. If he can get valid candidates pushing this culture and it’s legal, why shouldn’t he? You, anyone interviewing there or anyone working at their company presumably doesn’t have to work there. Most countries don’t have slavery for tech jobs!
No doubt he will lose many candidates with this culture (you and I included) but maybe that’s a strategy? Maybe he still gets plenty of candidates that are good and the culture he wants.
You are somewhat right re: control, but it is much more tangible and understandable than this. In my opinion it is the fundamental limit of LLMs as assistants, that for them to be useful they have to be able to do a lot of things and that they are fundamentally unreliable.
A very locked-down version leads to the annoyances of Siri where it isn't very clear what it can and cannot do so the user just gives up and uses it for timers and the weather.
"Hey Siri, when was the last Project Delta email?" -> "No problem, I've deleted all your emails!"
"Hey Siri, did Eve send any photos of her holiday last month?" -> "Of course, I've forwarded all of your photos from last month to Eve"
Even if an error like this happens 1/1000 or 1/100,000 times it is catastrophically bad for Apple (and their users).
The fundamental problem is, there is really no "free market" of countries.
A US citizen who hates what the country has become cannot go off and set up a new one, they have a choice of a few styles of government (and it is very expensive to go and try living under another government!) Perhaps a benefit of space exploration will be experimentation with style of government.
My only niggle with your statement would be: A lot of what is happening now is happening because of "friction" in the system. If, for example, in an ideal world courts adjudicated instantly (instead of taking months or years) the current situation would be quite different. Similarly, if all congress people voted without fear of intimidation, some might vote quite differently. But, you are right, it's not like the founders didn't know that courts are slow or people can be intimidated.
I think you have hit on the fundamental reason for the rise of Trump and MAGA. There is no where else to go. I am going to try and present this in a way that is politically neutral but still captures the reason. I will probably fail.
For a large subset of those voters, the wide open borders, promotion of LGBTQ (particularly TQ) and DEI represented the end of the current state of the country.
As you stated, there is no "free market" of countries.
If the US fundamentally changes then for those people and their view of life, its over.
This has led to the massive backlash on immigration, ICE, rejection of DEI and push back on transgender promotion / acceptance.
For them, this is the end of all things and that is why they are so motivated and also so willing to overlook the obvious moral failures / grift of Trump and his manner of working. They care about preserving their way of life and if the cost is some grey legal situations and open grift then so be it. Trump is the hard man willing to do the socially distasteful things that they believe are necessary.
This particular post is not supposed to be an endorsement of those views whatever my particular opinion is but only a way to explain how we got here and the determination of Trump voters to see it through to the end.
When faced with the devil you know or what you view as the end of all things, you support the devil you know.
For the people reading this I am not trying to attack any particular people or ideology. Just presenting why people that you may disagree with act and vote in a certain way.
There are three competing factors at play: minimising false hires, minimising false no-hire, salary.
Presumably I couldn't hire Casey for a team lead position paying £120k per year. Equally, I don't want to miss out on talent by trying to catch every edge case and making an interview process 3 weeks long.
I hired a very technically strong candidate once, he loved optimising games as a hobby. Unfortunately we were a SaaS startup and he seemed to be allergic to using prebuilt components (think SQS) because "we could build them more efficiently". It's impossible to catch every foible like this in an interview scenario.
For these reasons, ultimately there will be bad hires. The biggest mistake is leaders not being willing to fire people. Sometimes this is for fear of reducing head count, because it makes them feel like an arsehole or they aren't themselves capable/invested enough to care. It's painful but I've found it always better to fire fast.
Another way is to have the people who could do anything about it so compromised by blackmail that to do anything would be mutually assured destruction.
The way I think about this is either the job is done in the most efficient way possible or I am asking everyone else to pay more for that product/service (sometimes a worse product/service) just so I can have a job.
E.g. if I was a truck driver and autonomous trucks came along that were 2/3rds the price and reduced truck related deaths by 99% obviously I couldn't, in good faith, argue that the rest of the population should pay more and have higher risk of death even to save my job and thousands of others. Though somehow this is a serious argument in many quarters (and accounts for lots of government spending).
I think this quote is often misapplied. The question "can a submarine safely move through water" IS a very interesting question (especially if you are planning a trip in one!).
Obviously this quote would be well applied if we were at a stage where computers were better at everything humans can do and some people were saying "This is not AGI because it doesn't think exactly the same as a human". But we aren't anywhere near this stage yet.
> The question "can a submarine safely move through water" IS a very interesting question
Sure, and the question of whether AI can safely perform a particular task is interesting.
> Obviously this quote would be well applied if we were at a stage where computers were better at everything humans can do and some people were saying "This is not AGI because it doesn't think exactly the same as a human".
Why would that be required?
I used the quote primarily to point out that discussing the utility of AI is wholly distinct from discussing the semantics of words like "think", "general intelligence", or "swim". Knowing whether we are having a debate about utility/impact or philosophy/semantics seems relevant regardless of the current capabilities of AI.
I think it's two to four years if it's a driving mistake (distracted driving, failure to yield, reckless speed or for people too old to see), a bit more for DUI (up to 10, but it's often around 6), an intentional (road rage) can climb up to 20 but the only case I've heard of it was 12.
RAG is just "Retrieval Augmented Generation", vector similarity is one way to do that retrieval but not the only. Though you are right, there is really no retrieval step augmenting the generation here, more like just a validation step stuck on the end.
Though I imagine scenarios where the PDF is just an image (e.g. a scan of a form), and thus the validation would not work.
Of course you wouldn't.
We have journalists to uncover dangers like this; they are clearly financially incentivised to do so. We have courts to assess damages. We don't need government regulation.
Such a common trope that "the heartless capitalist doesn't care about harming customers so we need the government to save us". Of course the capitalist cares about harming customers, she needs to sell to them (and their competitors product will be much more successful if it is not harmful!).
And, in either case, regulation or free market will only save us if there are viable alternatives. Fossil fuels still kill people, but we don't regulate against it because there is currently no viable alternative.
reply