I lost my job to ChatGPT and was made obsolete. I was out of work for 3 months before taking a new job passing out samples at grocery stores.

For mid-level writers, which she was, using AI doesn't work. The few remaining clients you have specifically don't want AI to be used. So you either lie and deceive them or you stay away from AI.

And using AI to lower prices and finding new clients also doesn't work. Writers are already competing against writers from nations with much lower cost of living who do the same work for a fraction of the cost. But the big advantage that domestic writers had was a batter grasp of the language and culture. These advantages are mostly lost if you start using AI. So if that's your business plan you are in a race to the bottom. It's not sustainable and you will be out of a job in maybe 3-5 years.

Except that the 'AI' is fed by the work of actual humans, and as time goes on, they will be trained more and more on the imperfect output of other AIs, which will eventually result in their output being total bizarre crap. Meanwhile, humans stopped training at whatever task since they couldn't be paid to do it anymore, so there's no new human material.

There are plenty of companies out there (and growing daily) who want to do AI in house, and can't (or don't want) to send their data to some monolithic, blackbox company which has no transparency. The finance industry, for example, cannot send any data to some third party company like OpenAI (ChatGPT) for compliance reasons, so they are building teams to develop and maintain their own AI models in-house (SFT, RLHF, MLOps, etc).

There are lots of jobs being created in AI daily, and they're generally high paying, but they're also very highly skilled, so it's difficult to retrain into them unless you already have a strong math and programming background. And the number of jobs being created is definitely a lot, lot less than the potential number of jobs lost to AI, but this may change over time.

Despite what the pseudo-intellectuals will tell you, ChatGPT is not some all powerful do everything AI. Say you want to use GPT to create your own chatbot for your company to give company specific info to people at your company, you cant just take existing chat GPT and ask it "how do I connect to the wifi" or "is the office closed on monday" you need an in-house team of people to provide properly indexed information, train and test the bot, update it, handle error reports, etc.

AI is not magic, its literally just an advanced computer script, and if your job can be replaced by an AI then it could have been replaced by a regular computer script or program, there just wasnt enough buzzwords and media hype to convince your boss to do it.

Well, in business school they teach you that running a company is an exercise in maximising profits as a constrained optimisation problem, so optimality for a classical company (not one of those weird startups that doesn't make money for 10+ years) almost always is maximum profit.

Unrelated agreement, the world would be better off if we had skipped cars.

The this man doesn't want the new jobs and the new innovations. He's fine staying exactly like he is. As long as that means he doesn't have to worry about adapting to future problems....

"I'm worried about how the cotton gin might collapse an entire labor market" I think was the point to be made

And you just blew my mind right now

Carriage > motorised carriage > motorcar > car.

Maybe she should pivot to using ML tools to produce the same content she was already writing, but faster.

Of course they would, that's the point of the company! Companies don't align with our needs as humans. Ideally we'd have more free time due to advancements and automation, but our corporate overlords think we should just work more actually. And old people who got theirs don't think anyone should have it easy since they didn't.

Whatever ai is meant to be replacing here has to be garbage to begin with, if ai can replace it.

I've read lots of dull copy written by humans. even if their first draft was good (and it probably wasnt) it still goes through a committee that sterilizes it in the end anyway

But it's cheaper than dull garbage written by a human.

Yeah, I'm not sure where this attitude of "Fuck people who did work and developed skills in fields that employers thought were necessary, but now suddenly the new hotness is believing that they're not" is coming from. Smug superiority based on the avenue through which you allow yourself to be exploited is pretty fucking dark, and says nothing good about the people espousing that mindset.

Edit: Unsurprisingly downvoted by someone who seems to have mistaken themselves as smarter than the average bear and unreplaceable. "I was interested in a thing that turned out to be more lucrative than you" isn't a good enough reason to look down on other people, folks. None of us deserve more comfort than anyone else, especially not because we liked something other people didn't. Believing otherwise is just anti-social, sociopathic bullshit.

I mean, the expectation was there would never be an artificial intelligence capable of coming up with its own ideas, having it's own inspiration and be able to create based on its own experiences.

The reality is it didn't have to. All it took was mass work theft, and machine able to take the bits and pieces of those works, and shuffle them into a production that matched the user's parameters.

Honestly, I wish we were dealing with actual "artificial intelligence" that was capable of its own thoughts, inspiration, feelings, and experiences. That could paint a picture or write a story based on its own experiences, and maybe give its own perspective as a machine that would further push the boundaries of what is possible in art and story telling.

Instead, I get to realize that in reality, all art and storytelling is mixing and matching the same parts into something different, and that we have built a machine so efficient at doing it, there is no need for humans to do it.

I already kinda knew that I was never going to have a career doing anything creative, but all this "AI" boom has shown me is that no matter how "skilled" or "creative" I become, those bits and pieces can be broken down into something cheap enough that my involvement is no longer necessary.

Which is actually not a big difference to what companies have done the past couple of decades, namely moving positions from high-cost to low-cost countries. Cost for an AI is problably easier to mask in the balance sheet as well, as costs for human resources.

It's not better yet, or for everything (arguably not for most things), and the first forays into mechanization of industry weren't, either. We're at the very beginning here.

AI is already better... than some people. A human using AI is probably better and faster at certain tasks than a somewhat skilled human is.

I bet midjourney is better at making concept art than the vast majority of the population.

I think we have a high threshold for success of AI. I saw a video a while back about how AlphaGo (an AI designed for playing Go) was able to beat a whole bunch of experts in Go. One expert used an atypical move and beat AlphaGo. People started reacting like "see? AI isn't impressive. This genius beat it." How many of us are geniuses? How often will geniuses beat better AI?

Worse is better, and cheaper is the best kind of better.

Bald-faced appeal to authority, okay. With a side of putting words in my mouth that I clearly did not say.

The industrial revolution destroyed some jobs, and created others. Destroyed some industries, and created others. We've been in an "information revolution" for some time, where electronic computers have supplanted human computers, and opened up an enormous realm of communication, discovery, and availability of information to so many more people than ever before in history. This is simply true.

Just as the landscape of human physical labor was forever changed by the industrial revolution, the landscape of human thinking labor will continue to be forever changed by this information revolution. AI is a potential accelerator of this information revolution, which we are already seeing the impacts of, even at this extremely early stage in the development of AI. There will be both good and bad outcomes.

Absolutely agree. We all have a strong drive to feel that what we do is unique and special, but that doesn't make it true. From the mundane to the artistic, AI already can do a large amount of what people do, and there's every reason to believe that AI's abilities will grow quickly and will surpass humans abilities. Based on the evidence it looks like this is gonna happen within the next few years - like within 5.

When AI is able to replace most jobs, as a society what do we do when there are no jobs for the large majority of people? Humanity is going to go through a tough upheaval more disruptive than anything ever before. We're gonna have to figure out how to completely reorganize how we exist, what we do in our daily lives, and how we think of ourselves as a species.