alpos

The manufacturers make the pilot training requirements in cooperation with the FAA.

https://www.faa.gov/foia/electronic_reading_room/boeing_read...

https://services.boeing.com/training-solutions/flight-traini...

"Pilot Error" is the most favored scapegoat in the aviation industry. It's always possible to say the problem occurred between "the keyboard and the floor", to use the analogous software catch phrase.

But that isn't good enough when the systems are supposed to be designed for trained humans to operation without making such errors. If operating the system is error prone, change the system.

In this case in particular, if the pilots were not trained well enough that was due to the fact that Boeing's initial training materials were inadequate.

I know that article wants to push the narrative that those pilots in particular were not good enough pilots to begin with, but that was not true before they flew the Max8. Flying the Max8 is what got them killed. Kinda weird that they had this supposedly inadequate skill set all along but it was only a deadly issue with the brand new plane. It may indeed be that Ethiopia needs to up their training discipline. And that's good to address. But it does nothing to absolve Boeing's part in those crashes.

https://www.businessinsider.com/boeings-ceo-on-why-737-max-p...

The FAA and Boeing have now established the new training requirements that should have been there from the start.

> "3. HIGHLIGHTS OF CHANGE The purpose of this revision is to add training requirements for Maneuvering Characteristics Augmentation System (MCAS), Autopilot Flight Director System (AFDS) enhancements, and additional Special Emphasis Training. Appendices 2 and 3 reflect applicable table updates from the Boeing 737-800 to the Boeing 737-8 and from the Boeing 737-8 to the Boeing 737-800. Appendix 7 was added to delineate ground and flight training necessary to operate the 737 MAX modified with Flight Control Computer (FCC) software version P12.1.2."

Post incident Boeing is doing some double speaking for PR and CYA. But the investigation already proved out that they were dodging the training updates so they could get the plane to market sooner. And two things can be true at the same time as well. Accident investigations are meant to uncover ALL of the systemic and specific failures that led to the crash. And that's a really good thing.

But again, everything thing even Boeing is responsible for, was done to avoid costly extended certification requirements for the planes and time consuming re-certification of pilots.

If regulations are leading to this kind of behavior, it's not enough to even blame just Boeing. We need to make the right thing to do, the easiest thing to do. Maybe that's not possible here, but I don't really see anyone trying.

Before I started flying, that was my imagination as well.

Small Aircraft engines are actually some of the simplest designs you will ever find in a four or six cylinder engine. They are essentially pegged at 1950s to 1970s tech level.

The main difference you will notice first in the controls is that they all still require the operator to manually control fuel mixture. Cars used to be that way too.

But having a carburetor with a manually controlled fuel inlet valve opens the door to the operator getting it wrong and running the engine either too rich or too lean, which can lead to pre-detonation and/or build up of carbon deposits in the cylinders and on the spark plugs. Both of which throw off the engine timing, wear out engine components faster, and can lead to cracking the engine housing, breaking a piston rod, or plain old sudden loss of power during flight when the spark plugs stop working.

And that's on top of the possibility that the venturi can freeze up, starving the engine of fuel and air on decent, even on a hot day outside. Which means the pilot also has to learn to use "carb heat" which is an inefficient solution that sends hot exhaust gas back through the carburetor.

The controls are just two levers, nothing complicated, but the ways one can mess that up are varied. They are also subtle mistakes, right up until you are suddenly losing power at the worst time.

The answer in the car world has been to switch to fuel injectors with computer monitoring and control.

However, even introducing FADEC systems has been a struggle in aviation, LONG after they were well proven in ground vehicles.

As a result, operating just the engine in a small GA plane is harder and more error prone than operating a car engine and the aviation world responds to that increased risk by adding even more regulation around maintenance requirements and overhaul times. Which sounds perfectly rational but there is little data behind the regulations compared to what we could have if most engines were computer controlled and monitored. And we wouldn't need to be so paranoid on maintenance hours if we could rely on issues being noticed before they become problems.

Additionally, this makes flying with those more basic engines a bigger risk because the pilot then has to know the sounds, vibrational patterns, and general "feel" of that particular engine in order to have a good chance at detecting any issues that might lead to a loss of power. But as we humans are well aware from experience making car engines, this is not good enough, there are hundreds of little bits if information that can be indications of the beginnings of an eventual and costly failure that, if noticed early and remedied at the shop, results in continued and more reliable operation.

We start our cars every day and drive to work without really even thinking about the engines. The computers have our backs there. And the vast majority of the time, if anything, the computers are a bit TOO paranoid. But that's a good thing because it means problems get fixed before you end up broke down on the side of the road.

It is even more important that the engine not fail suddenly in flight because then you are forced to land wherever you can, even if that's a farmer's field or a highway. Very dangerous for the people in the plane and somewhat dangerous for people on the ground too.

One would think that the importance of detecting issues BEFORE takeoff and as early as possible during a flight, before the engine actually starts losing power, would make us want to use every tool at our disposal to operate, monitor, and maintain those engines as best as possible. But this is not the case, largely due to the way in which aircraft engine manufacturing and certification are regulated.

The underlying problems here run much deeper in that the root cause of the issues has more to do with the shenanigans Boeing and any other aircraft manufacturers end up having to do in order to avoid having to certify a whole new aircraft and especially to avoid needing to obtain a new type certification.

Both cases are absurdly expensive and time consuming to the point that any new aircraft designs are basically experimental or dead on paper in terms of being able to sell the plane at a price people or even companies can actually pay.

In this case, my understanding is that, the new engines are a bit bigger which necessitated a change in mounting position relative to the main body in order to maintain ground clearance on takeoff and landing, this should have meant larger landing gear sitting a bit higher and/or a slightly different wing shape and mounting position on the body.

However, any of those kind of changes would have moved the Max8 into a new type certificate which would then require all pilots flying it to certify on that type in addition to additional flight testing and manufacturing certifications. All of that would move the cost of the plane well above the point where the efficiency gains would make it worth buying and delay production to the point where there would no longer be a market for it. So there would simply be no Max8.

And one could say, "well that's business, Airbus beat you to it, Boeing should lose then". And that would be fine as long as we're willing to accept a world where the US produces no commercial aircraft at all and airline ticket prices and accommodations continue to get even worse. Are we good with making travel by air something only the rich can afford?

So Boeing went with extending the engines forward and slightly upward which necessitated more down force from the horizontal stabilizers to balance the increased weight forward of the center of lift and introduced a potential disruption of airflow over the wings in a high pitch up situation.

Changing the size, shape, or position of the horizontal stabilizers or anything else that would address the airflow issues would also run into certification cost issues. Therefore Boeing settled on an active control solution which they integrated into the existing flight control software (poorly) and found creative ways to convince regulators that nothing much had changed in the software such that the whole design could be built at a price point and time to market that would actually make the project worth doing at all.

This kind of creative engineering of what you can get past the regulators is the dark side of any over-regulated industry and continues even in the re-certification process for the Max8 https://www.theverge.com/2020/12/18/22189609/faa-boeing-737-...

The reasons for it are pretty straight forward and can be seen as bad actors being bad or simply the result you are bound to get when you regulate without accounting for the reality of how people and businesses work. You can't just say "safety is our top priority" and continue to add layers of regulation until it's impossible to do anything without manipulating or bypassing those regulations. Doing so will counter-productively NOT get you safer designs.

What it will reliably get you is a shift in the focus of efforts from engineering the safest and most efficient aircraft possible to getting ever more creative about designing things you can get past the regulators. You don't want smart people dedicating large amounts of time, energy, and money to efforts to manipulate regulators.

We the public would be much better served by an FAA that regulates with safety as the top priority AND an eye for making it possible and profitable to actually introduce new designs and new ways to manufacture things.

Right now it is simply impossible to start a company with the goal of introducing a new aircraft or aircraft engine to the US market. It costs astronomical amounts to get to the point where you can sell your first plane and separately astronomical amounts to get to where you can sell your first engine. Even for single engine, fixed gear, small and light, GA planes.

This is why almost no normal US citizen can afford to fly personally, even just occasionally for fun. It is also why GA planes still use leaded gasoline and burn around 10 gallons per hour. It is also why "new" aircraft are all actually >95% similar to existing designs (even if that makes them inefficient), always use existing decades old engines, and still cost hundreds of thousands to millions of dollars, making them priced out for all but the wealthiest potential customers.

The only possible route for any innovation in flight has been Experimental airworthiness certificates. And you can sell a few planes that way; but once you have a known good and tested design, you are still stuck in the Experimental category indefinitely because you will never sell enough planes to afford a standard category certificate.

Best of luck to the flying car startups but no pilot in the US expects them to have a snowball's chance in hell of making it to the point where they are profitably selling a genuinely new type of aircraft.

Why is this better than just using Dart/Flutter?

The introduction docs also claim "In Electron and NWjs, you have to install Node.js and hundreds of dependency libraries. Embedded Chromium and Node make simple apps bloaty. Neutralinojs offers a lightweight and portable SDK which is an alternative for Electron and NW.js."

Yet the installation here starts with NPM and depends on Node.

Sorry, but I don't see a significant difference between this and Electron just from the comparison chart linked in the docs. https://github.com/Elanis/web-to-desktop-framework-compariso...

Save for the fact that Neutralinojs is not listed as having any support for automated testing where Electron does.

The data there shows smaller and faster builds on empty apps but that doesn't count for much since no one ships empty apps.

Needs a fair comparison against its claimed competitors using app examples that at least mimic real world apps. For example, if building the typical kind of application people would build in Electron still results in a smaller deliverable and much faster build, that would start to make a case for a better developer experience.

Though I would guess that most developers would not switch and learn a new framework just to get slightly faster builds before considering test time. Integration test run times tend to take up most of the time in any build process.

Forgery occurs when someone makes a copy of an original work and tries to sell it as if it were the original or as if it were a variant created by the original artist.

You can sell a copy of anything not covered by copyrights. You just can't lie to people about what you are selling.

App developers have to deal with this every day. And it's call a free market in that context, I don't see the difference between that and art in this context.

People straight up clone every good app that shows up on the app markets and you might as well make and sell templates for any good looking, well designed web site you own because someone is going to scrape it and do that for you if you don't. That is what we have to deal with in the software world all the time and any complaints there simply get a shrug and a "welcome to capitalism" from the more experienced developers.

If it were stealing, the artists making the accusation could make their case in court or perhaps the companies could since they own the rights, but that also means the comic artists actually have no grounds for complaint, they sold their rights. You don't get to complain about the use of something you don't own anymore.

I can understand the shock some of the comic artists must have had when discovering later in life that Lichtenstein had done this; but, their anger should be directed at the companies who gave them a bad deal as well as at themselves for accepting that raw deal and never looking for something better.

For any actively working comic strip artists though, if you are aware someone is making 10x, 100x, 1000x off of very similar pieces based on your original work, you have no excuse to keep doing what you're doing.

That person has demonstrated that you are wasting your time and giving up the rights to your work for pennies on the dollar compared to what your work could be worth. If you want to make millions too, the obvious thing to do is make a full sized image of anything you make for the comic strips and sell those in art galleries. If you can't do that under your current contracts, drop those contracts.

You can always out compete copycats. You're the original, so you always have an edge there. And if you can't out compete Lichtenstein, then you must admit that he is doing something other than simply copying you. There's something he's got that people see as more valuable than just the image itself. Figure out what that is and do that too.

By the way, even if your full sized images don't sell well, the fact that you are selling in art galleries gives you grounds to sue the guy if he keeps trying to copy your work.

There were always ways to stop this guy or replicate his success, if any of those comic artists really wanted to. But they didn't start complaining until he was making a bunch more money than they were. That is just raw jealousy.

Seems like there's a bug in that system, it was discovered on accident, and now there is a bot that exploits this flaw.

The fix will most likely have something to do with requiring that citations use primary sources, not just any article on the internet. In then end state, Wikipedia will be much better for it.

Probably add one caveat to the first case, if the bot reliably produces libels the same person when various users ask the same or similar questions, then there is damage and most likely whoever is running the bot is liable.

Well, at least you are honest about not really caring about anything other than profit.

I'd love to see what makes you think think almost everything that matters to most people for most of their lives is irrelevant to the question of how people live and work on a daily basis. Please elaborate if you will.

Since you disagree with my characterization, I'd also be happy to consider what you think I have misattributed there.

Overall the point is simply that you are not working with a steady state system here.

Whether one mode of working confers a competitive advantage probably has more to do with how much effort has been put behind streamlining that process. If that is indeed the case, then you might find yourself saying this about any change to any process that doesn't just wipe the market and create a whole new competition environment.

There's no good reason to throw our hands up and say there's nothing we can do to make one or the other situation work better, especially when the benefits are great, numerous, and include increased productivity overall.

So fix those issues? Working in offices always had issues as well, some of those issues have been addressed over time though effort on the part of employees and managers. Nothing is without trade-offs and side effects.

The real question there is: Are the minor issues associated with working remotely so bad that it is better to

  force everyone to risk their lives on the roads every day (including in snow and other bad weather which exacerbates road death rates 100% purely because of employer demands to come in anyway)

  destroy the environment with all of that unnecessary exhaust,

  give asthma to their neighbor's kids,

  harm parent/child relationships due to work schedules being out of sync with school schedules and work/commute demands leaving little time for parents to properly interact with their kids,

  pass flu and cvoid around to everyone,

  cause unnecessary deaths among their parents and grandparents,

  drive up housing costs in specific areas causing homelessness as well,

  and otherwise add layers of stress and unhappiness to every person's daily existence?

Put the other way around: Are all of the personal, professional, societal, and environmental negatives that directly result from people being forced to commute to and work in an office every day actually worth the supposed productivity and/or collaboration gains due to office related friction points having already been solved over time? (btw they really haven't been solved for many people either, see: the ongoing harassment and diversity issues at most companies)

Or might it be better to just solve the few issues with collaborating in a remote work setup like we already did for the office?

>"By arbitrarily forcing return-to-office without providing data to support it and despite clear evidence that it is the wrong decision for employees, Amazon has failed its role as earth's best employer," one employee wrote, according to the report.

Both this and open floor plans were never about data. This is about control. Nearly all employers, especially all of the oldest and largest will steadily pull their workforce back into offices where the bosses can satisfy their egos with the notion that seeing their people work makes them in charge. They don't just want to get things done. They want to see the buzzing activity of the office and feel like they are in control of it.

And this will happen in spite of any protests from the employees. By now most companies, especially tech companies, are well practiced at forcing undesired changes on large groups of people. They even have most of their customers actually paying to be spied on. Something no one would have accepted even 15 or so years ago.

So long as the company is arranged as a hierarchical power structure, this will be the way. That's the whole point of talk about things like "culture" as well. The reader may take note that, between companies in a supply chain, there is no talk about or desire to control culture even though those companies clearly depend on each other to do their business at least as much as they depend on their respective employees.

Only within the company does anyone seek to control and define "culture". That is because setting company culture is means of controlling employee behavior. And it only applies to those who signed employee contracts. Consultants, general contractors, support staff, and vendors are almost always treated more as real and self-determinant humans than the employees the company leaders seek to control.

In short, your boss doesn't see you as a peer or as a fully capable rational agent. They see you as a ward under their care. This is a side effect of the power imbalance created by the structure of the company.

Isn't it strange that almost everyone on earth desires to live under a democracy more than a dictatorship but when it comes to the interactions that will dominate most of their days for most of their lives, somehow people are okay signing up to work under a dictatorship?

...And anyone still under the illusion that Amazon is or was ever earth's best employer might want to try talking to the delivery drivers and warehouse workers.

Belligerence is an intention. Bullshitting is a technique.

The bot is incapable of intention so the entity here is still the human who prompted it. What the bullshit it produces is used for depends on their intentions.

"We have also secured the services of Equifax to provide identity monitoring services at no cost to you for two years. Below please find information on signing up for a complimentary membership to Equifax’s identity monitoring services, including key product features"

Really PayPal? You mean this Equifax?

https://en.wikipedia.org/wiki/2017_Equifax_data_breach

The security breach problem altogether has been turned in to a source of extra profit for the credit reporting companies whose faulty systems, lack of data verification, and resulting poor data quality are the root cause of these issues in the first place.

The problem has never been "oh no someone stole your identity!". Identity can not be stolen, only temporarily misattributed in the eyes of one party or another.

The actual problem is that companies like banks and credit card issuers don't actually know who they are dealing with and don't want to put any effort in to verifying that. Using just a few pieces of already available, static information about a customer was never enough to say you actually know who your company is signing a contract with.

Calling it "identity theft" when someone feeds your faulty signup process incorrect information is merely a means of passing the problem with your systems off to the customer. Now we're going as far as pretending the root cause of this stupidity can also sell us the cure? Nonsense.

If anyone ever "steals my identity", I'm going to sue the credit reporters for libel. They have admitted before that some 70% of their records contain inaccurate information and they know full well that a few bits of static information is not enough the identify an individual among hundreds of millions. They also know they don't do anything to establish that any new reports coming in from partner businesses are actually real and accurate. Yet they still peddle that inaccurate information to third parties. That is the same as if a journalist made up a story saying you did some shit you didn't do. They are knowingly giving out information about a person that is either provably false or which they have no good reason to believe is true in a way that can cause financial harm and loss of reputation. Classic libel.

Further, if some business reported that I took a loan from them when I didn't, that isn't my problem, it's theirs. They failed to adequately verify who they were dealing with. Not my problem. Reporting that inaccuracy to a third party who then broadcasts that to a bunch of other companies, is also libel.

They can make a case that they didn't know for sure that their claim that I owe them was false at the time, but they actually never had a good reason to believe it was true. Static information alone, especially that which we already know is in circulation among public entities, is not a valid way of identifying who you are dealing with. Pretending that it is anyway is just bad business.

People need to hold these businesses accountable for the damage their faulty system designs are causing.

I was not suggesting the bot was telling the user to go make a PR to an open source framework but rather that they could create a library that contains those functions if that was the logical thing to do. Which is why I asked if that actually seemed the longer term right thing to do.

While I can easily agree that Copilot is probably the better tool for such questions, it is not clear from the parent comment whether their prompt to ChatGPT was asking to create code to do X or create code to do X using only existing publicly available libraries.

It's not immediately obvious that the bot failed to understand the question or that the answer was an example of the bot failing to understand the programming language. It could easily be that the user had an implied expectation of usability in a context they did not give to the bot.

That scenario is more like you asking a random person on the street, who happens to know Y framework, how to do X in that framework. Your coworker can be expected to get that you are looking for an answer that gets your current task done faster than you would be able to do without their assistance. The person on the street could not reasonably be expected to get that unless you give them that context.

Saw that, and yea, that's totally a fall-back cop-out type answer. I was pretty much just questioning everyone dismissing the whole category of answers like the R script one.

To me there does seem to be some nuance to here that's worth noticing. Some examples of this type of response are indeed too cheap and can be chalked up to lack of training data or something.

But in other cases it's actually not immediately obvious whether the answer the user got was their fault for not specifying that they are expecting code that works without additional supporting libraries.

A language model can't reasonably be expected to understand an expectation of usability or fitness for purpose in a context the user didn't specify.

Agreed. That does seem to be an example of a language model failing to understand the context in which the question was asked.

The user was implicitly expecting code that would function when executed immediately and as written with no additional supporting libraries included. This is different from code that would function correctly when executed after having downloaded relevant existing packages. Which is different from code that would function if executed along side additional supporting code from private libraries the user might not have access to. Etc...

Yet any of those answers fit for the same prompt, "create an R script that does such and such". The bot's lack of insight is on the likely intention behind the prompt rather than on the requested language. I'd say if it produces any code that fits the syntax and grammatical structure of the requested language, that's enough to say it understands the language.

Actually it's the other way around.

Those pushing the concept of "AI-Free" have yet to nail down what amounts and types of automation may be used in the production of a thing before it can no longer be labeled "AI-Free".

On the other hand, "Organic" is very well defined at this point. In the US there is a whole legal framework around labeling foods, drugs, and cosmetics.

https://www.fda.gov/food/food-labeling-nutrition/organic-foo...

https://www.ams.usda.gov/about-ams/programs-offices/national...

There is even a training and accreditation system in place for qualifying people as certifiers of organic practices as well as an application and review process for farms and manufacturers to be certified to use the label.

https://www.ams.usda.gov/services/organic-certification/beco...

Responsibility and ownership always lies with the humans. Even supposedly 100% AI generated content is still coming from a process started and maintained by humans. Currently also prompted by a human.

The humans running those processes can attempt to deny ownership or responsibility if they so choose but whenever it matters such as in law or any other arena dealing with liability or ownership rights, the humans will be made to own the responsibility.

Same as for self-driving cars. We can debate about who the operator is and to what extent the manufacturers, the occupants, or the owners are responsible for whether the car causes harm but we'll never try to punish the car while calling all humans involved blameless. The point of holding people responsible for outcomes and actions is to drive meaningful change in human behaviors in order to reduce harms and encourage human flourishing.

In terms of ownership and intellectual property, again the point of even having rules is to manage interactions between humans so we can behave civilly towards each other. There can be no meaningful category of content produced "100%" by AI unless AI become persons under the law or as considered by most humans.

If an AI system can ever truly produce content on its own volition, without any human action taken to make that specific thing happen, then that system would be a rational actor on par with other persons and we'll probably begin the debate over whether AI systems should be treated as people in society and under the law. That may even be a new category distinct from human persons such as it is with the concept of corporate persons.

I have to wonder, were any of the the hypothesis' in those papers plausibly viable areas of inquiry?

Perhaps it could be useful if they train the bot to identify cases like this and state that no such references exist but also provide a thesis or suggest a line of study that would produce such a reference.

This take strikes me as a little off. Programming languages are language. Unlike natural languages they are also based on context-free grammar. So an understanding of programming languages should actually be easier for even a general language model to incorporate than natural languages.

We can expect a bot like this to not really get context clues in natural language, although they seem to be getting better at that, but context is not necessary to have a true and functional understanding of a programming language. That was the point of creating such languages.

Using an API that doesn't exist but logically should once the use cases are demonstrated is not an example of lacking understanding, it is an example of advanced insight. A human might have invented the necessary functions inline with the rest of the project but if they are expressing functionality that is commonly applicable, then a common API for those functions is what the humans would eventually converge upon to clean up the code from the initial inline implementation, making it more consistent and readable.

That seems more like misplaced expectations. Someone may have given you the impression that copilot was supposed to do things like that where that expectation seems to not be present for you in relation to ChatGPT.

However, as far as I know, the OpenAI team has not made it a goal to have ChatGPT only produce functional code using existing APIs. So I'm not sure we can call that an incorrect answer based on context.

If the API it demonstrated using logically should exist, it seems like the right answer is still to just go create it.

This strikes me as a distinction without a difference. Content creation is not even close to the same as GMO or processed foods. People are justified in being concerned about how food is made precisely because there are meaningful differences in the nutritional value and trace chemical content in the end products. And that matters because we don't fully understand the role those factors play in long term human health. Created content does not share similar distinctions based on how it was produced.

At the level of our personal experience of a story, image, or sound it does not matter if the content was human generated or even if the stories are real; they either elicit an emotional response or they don't. This is why we can enjoy fiction and art in the first place. It's also why a given piece of art can mean different things to different people.

Deliberately cutting yourself off from the meaning a piece could have for you just because you aren't yet sure that a human produced it is unnecessarily specist and only acts to your own detriment. Should one likewise refuse to engage with works that had assistance from the artist's pets?

With regard to news stories and other informational content, if two stories about the same event are both accurate, timely, informative, and/or insightful, it does not matter if one was written by a human on a typewriter and the other written by a bot after having been prompted by a human. The present state of the art still requires some human editing to hit those marks but we're quickly approaching a time where there will not be any detectable differences.

This idea that people deserve to invent a new category of things to be prejudiced about seems silly and unhelpful to me. Why bother inventing arbitrary distinctions just so people can play favorites and be judgemental about it? People are already free to judge content based on relative merits. That seems like enough to me.

And it's important to account for the fact that the end goal with these tools is that there will be no differences between human and AI generated content. The teams working on these things have already made considerable progress to that end, it's not hard to imagine the next couple of iterations actually achieving that.

So anyone trying to create these arbitrary labels is staring an uphill battle for a fundamentally unhelpful end state. We really don't need to put that kind of energy into reinforcing people's desire to feel superior. We have more than enough elitism/triablism/racism/otherism around as it is.

> I've already run into scenarios where ChatGPT generated code that looked perfectly plausible, except for that the actual API used didn't really exist.

So the next question has to be: Was this still the right answer?

I've personally had plenty of instances in my programming career where the code I was working on really needed functions which were best shopped out to a common API. To avoid interrupting my flow and to better inform the API I'd be designing for this, I just continued to write as if the API did exist. Then I went on to implement the functions that were needed.

Perhaps the bot was right to presume that there should be an API for this. You might even be able to then prompt ChatGPT to create each of the functions in that API.

The context is probably what's throwing you off then. I took bilsbie's comment to be a context expansion here. Imagining how this general strategy of censoring what's actually out there plays out in the long run and wider context of how all of humanity deals or doesn't deal with what human minds are actually capable of producing.

Also note that this wouldn't just be about a bot producing more bad stuff. ChatGPT also, or even mostly, answers questions and provides information. If bots like this do not even know about the filtered out content they cannot, for example, give an accurate or original answer to a question like: "Is nazi-ism on the rise?" The best they can do is regurgitate news posts, pundit commentary, and existing research articles. That might be good enough, but if a bot were able to see all the content out there it might even give a statistical summary of everything on the web and comment on the rate at which pro-nazi web articles are published compared to articles of all other types. That would be a useful way to shed some daylight on the issue.

If most of our efforts most of the time are spent merely filtering out the trash, then we're still just ignoring pretty big issues. The strategy as applied here isn't exactly wrong for what OpenAI is trying to achieve but it still ends up being another example of people being more willing to spend a lot of resources pretending humans are better than they are than we are willing to spend a lot of resources fixing the underlying issues that make some people so screwed up to begin with.

This new ml model they are building will be used to filter out child porn, among other things, from the output of ChatGPT, so users of that bot don't see it. It seems their ideal case is to reliably filter it from the input as well, so even the bot doesn't know about it. Which is fine for what they're doing, so they'll probably stop there.

However, once you have something like a reliable detector for human misery and the warped human minds the produce it you're actually well on your way to developing a tool chain that could help authorities and mental health professionals identify and track down those warped mind as soon as or possibly before they cause harm.

I'm not confident I know the whole field well enough to say no one is doing something like that but it does seem like the preferred option for most all AI teams is to spend $200k filtering out the problem rather than taking the time to build something that reduces or eliminates the root cause. And that stance is reasonable enough at the team/company levels because you'd have to do the former anyway even if you also do the latter.

So of course it's not OpenAI's job to do that and this isn't a moralistic judgement on the teams building AI in general either. Humanity just isn't interested in solving those problems for real in the same way we are interested in a bot that can write code or tell us interesting stories. We want to know the police are on it in some capacity, but there's not likely to be enough investment in it to actually solve the problems any time soon.

In the long run, being increasingly good at filtering this stuff out tends toward a future where most people can live their whole lives without ever becoming aware that these awful problems exist and possibly happen in their own cities. And if most people are never aware of a problem, it's not likely to ever get enough attention and funding to be solved.

So that is the sense in which I can find agreement with bilsbie's notion that merely filtering out the bad stuff humans produce can be more toxic than letting people see what humans are really like. Though again, it's hard to blame any one team for just wanting to filter that stuff out.

---

And there is actually at least one unrelated other way to look at bilsbie's comment so you might ask them what they meant as well. The other way I can think of is asking who decides what's acceptable content? And then expand the context on that to a hypothetical world where stuff like ChatGPT is ubiquitous. There are some obvious cases we'll all pretty universally call bad, so we don't mind someone training the bots to filter that out. However, there's other examples like a lot of horror books/films or even mystery/detective novels and such that push the boundaries in order to get people to think, if those get caught in the filters too, then we're doing humanity a disservice by effectively banning/digitally-burning books/articles/web-sites. And let's not forget, if people are good at building such filter systems for all the bots you can bet governments with a eye for controlling thought will be interested in installing some filters of their own.

No one is suggesting we should embrace the worst. Only that ignoring it at every chance instead of ever properly dealing with it isn't the most helpful thing if we actually want to follow that desire to go beyond.

> If a business derives $X profit from a laborer and agrees to pay a given % of $X to the laborer as compensation, it should not matter if the laborer is in a high cost or low cost of living area

That would be true if that were deal the business had with their employees. However, you made up the percentage thing by observation and are working from the assumption that because you can represent a wage as a percentage of a profit margin, that means the business agreed to share a fixed percentage of profit with the employee. This is incorrect reasoning.

Businesses generally don't contract with their employees to share a percentage of revenue or profits, with the narrow exception of commissions for people in sales. With regard to developers and almost all other employees, the contract both parties agreed to is almost always a trade of specific amounts of time for specific amounts of money and possibly some flexible additional benefits. And to the employee's benefit, the amount of pay does not flex based on whether the company is even producing a profit from the sale of the the products of their labor, only how many hours they put in. Caveat all manner of fuckery in labor relations, of course. Which is a giant caveat, I agree in advance.

However the fundamental agreement is still time for money without the employee having to worry whether the products of their labor can be sold for more than their hourly wage, that's the business's problem to deal with. The employee expects their paycheck to show up on time all the time, or the deal is off.

With that sort of agreement, it is absolutely expected and reasonable that the business keeps any profit from selling the product of the labor after the agreed upon fees for said labor have been paid. That was the deal.

---

If you'd like a different deal for most people, as I would love to see more of in the world economic future, then most people need to change over from needing a job to needing some form of income to support their long term future growth as they live currently off of savings.

Instead of expecting someone else to ensure there is a buffer of money to pay you so you can pay your expenses, you will have to keep that buffer yourself and be okay with the fact that sometimes you will get paid a lot and sometimes not so much, but you will then be getting that percentage of the profit you wanted and it will be totally reasonable to expect the same percentage regardless of what the cost of living is.

You also won't have to worry about layoffs as much because the company's costs now also scale directly with it's profits, so there is no need to cut head count when things get really bad and the company's buffers run low.

But again, that also means you will be responsible for keeping a big enough buffer yourself to deal with the fact that your income fluctuates with both short an long term market swings. Otherwise all you have done is trade the annoyance at seeing the company profit massively from your fix-time-cost labor for the annoyance of being broke even though you have a job whenever the market takes a shit.

Most people still don't want to deal with that, even in super wealthy countries like the US. In fact most people in wealthy countries actually scale their own living expenses in step with any increases in their income such that they can be living paycheck-to-paycheck even at $200k+/yr.

So the balance of things present day is that most people take full-time or even part-time contracts as employees and hope they don't get fired at the wrong time in life.

This is possible to change though, for yourself and for others if you like. So if you think your current deal is bullshit, feel free to change the deal.

So long as developers want to be paid by the hour rather than paid for a specific result, I would say this is fine too. The problem everyone is waffling on here isn't the competitiveness of the wage, it's the gap between the value of what the labor produces and the value offered to produce it.

But you can't escape that gap so long as you're fundamentally trading time for money.

If you want to capture more of the value of your work, you become a contractor and bid firm-fixed or cost-plus contracts instead of labor-hour or time-and-materials.

If you want to capture even more value than that, you work on building business systems and selling a valuable product or service rather than spending your work hours doing the tasks themselves.

Whether you're in Kenya or the US and whether you're doing undesirable or highly desirable work, it is not a moral failing of the business you are working for that they did not also take the time to ensure that you could have become a contractor if you wanted to or that they did not ensure that your economic situation allows sufficient market opportunities for you to quit and form a business if you wanted to.

I tend to agree with this notion.

Instead of dealing with the fact that humans do and produce tons of awful shit, we just do our best to ignore it and pretend human nature is better than it is.

This is delusion on a grand scale and it's hard to find examples in the historical record of good things ultimately coming from lots of people acting in the world while being divorced from reality.

It's worth noting that OpenAI actually paid around six times that amount for this work.

It's also worth noting that Sama actually did provide for their employee's mental health. They didn't do a good job of it but trying and failing at a thing doesn't quite qualify at the level of disgust imo. Especially not for the company that was outsourcing this.

It's also worth noting that even if those of us in the peanut gallery would not trust the word of either OpenAI or Sama, it is not unreasonable for OpenAI's team to have trusted that Sama was doing what they said they were with regard to providing for their own employees mental health.

That's not proportional, by design even. The whole game here is to do a small amount of work and multiply the gains by an infinite number of results from people using the bot.

That's always the game with any kind of technology. So no, having a few people experience something bad so billions can avoid experiencing anything bad ever does not cancel out.