First off, this argument completely sidesteps any ethical concerns with AI, so if someone is opposed to using it based on the fact it's stolen all of its training data or because of the effects AI data centers have on the world then there's really nothing persuasive in here.
Second, even if you don't have an ethical opposition to AI, I feel like you're applying your argument way too broadly. Sure, I can see how there's not much of a functional difference between "a PR department wrote this statement for me, but I'm the one delivering it and stand by it" vs "AI wrote this statement for me, but I'm the one delivering it and stand by it," but you completely disregard any artistic writing tasks. Even without the ethical concerns, I still want to read poetry written by people, not AI. I want character dialogue, short stories, music, etc written by humans. Your examples really only point to "you should be fine with computer code and boring writing tasks in the corporate world being taken over by AI" but are arguing that people should accept AI being used for any and all writing
I also think that the choice to create or consume entirely human-created content is completely fair. The way you’re describing it, it sounds like you’re thinking of cases where the writing itself is the point, the act of artistic expression by the writer, as opposed to service writing, where transmission of the idea is paramount, not the writing in itself.
Regardless though, this is a personal decision. Where I draw the line is making that decision for others. If you only want to read things entirely created by humans, great. But others may find utility in using AI to assist in the creation of writing to express ideas, summarize data and so forth. I believe that that is also okay.
I agree giving people the choice to engage or not is best at this moment, but that's why I'm in favor of AI disclosures and don't have a problem with Anthropic's water mark. I see it like a restaurant marking what menu items contain meat or are vegan friendly. No one's forcing people to be vegan, but we should disclose the necessary info so people can make their own informed choices
This was a puerile read. Setting aside the massive ethical problems with AI training data collection, copyright, and existing IP law, setting aside the massive energy costs and concomitant resource scarcity and environmental impacts, AI was not a tool created to unlock the creative potential of millions. No one is spending a trillion+ dollars of capex so you have an easier time writing an article.
Companies are spending a trillion+ dollars to not need expensive workers. AI exists to give capital access to skill, while denying skill access to capital. For all your examples of technologies superseding more laborious processes, you ignore the fact that they all pretty much *improve* on the former labor. I don't have to worry about my sloppy arithmetic mistakes when I use a calculator that is much faster and more accurate. My dishwasher can tolerate water far hotter than my hands can, enabling a far more sterile clean than I can achieve without it (and it uses far less water than I do when washing by hand). The skill I've given up is significantly outweighed by the virtues of the new substitute.
That is not true with AI. Coders are losing skill, and losing the experiential understanding of the code they're still responsible for. And the flood of insecure AI slop apps belies any argument that what we're giving up is worth what we're getting in return. And as far as writing? AI continues to beclown writers putting trust into the output of AI. If accuracy and trust mean anything to the writer or the reader, generative AI cannot be part of the process.
I noticed a problem with one sentence: "Why is your concern." - shouldn't there be a "?" at the end of it? Given the subject of the article, I'd hope for "perfect" proof-reading of its contents.
I could not agree with this any harder.
First off, this argument completely sidesteps any ethical concerns with AI, so if someone is opposed to using it based on the fact it's stolen all of its training data or because of the effects AI data centers have on the world then there's really nothing persuasive in here.
Second, even if you don't have an ethical opposition to AI, I feel like you're applying your argument way too broadly. Sure, I can see how there's not much of a functional difference between "a PR department wrote this statement for me, but I'm the one delivering it and stand by it" vs "AI wrote this statement for me, but I'm the one delivering it and stand by it," but you completely disregard any artistic writing tasks. Even without the ethical concerns, I still want to read poetry written by people, not AI. I want character dialogue, short stories, music, etc written by humans. Your examples really only point to "you should be fine with computer code and boring writing tasks in the corporate world being taken over by AI" but are arguing that people should accept AI being used for any and all writing
I actually agree with a lot of your points, however I think there’s some nuance to be had here. I do think that there are valid ethical questions around AI, in fact I’ve written about that here: https://ageofcomplexity.substack.com/p/faster-and-cheaper-moral-culpability.
I also think that the choice to create or consume entirely human-created content is completely fair. The way you’re describing it, it sounds like you’re thinking of cases where the writing itself is the point, the act of artistic expression by the writer, as opposed to service writing, where transmission of the idea is paramount, not the writing in itself.
Regardless though, this is a personal decision. Where I draw the line is making that decision for others. If you only want to read things entirely created by humans, great. But others may find utility in using AI to assist in the creation of writing to express ideas, summarize data and so forth. I believe that that is also okay.
I agree giving people the choice to engage or not is best at this moment, but that's why I'm in favor of AI disclosures and don't have a problem with Anthropic's water mark. I see it like a restaurant marking what menu items contain meat or are vegan friendly. No one's forcing people to be vegan, but we should disclose the necessary info so people can make their own informed choices
This was a puerile read. Setting aside the massive ethical problems with AI training data collection, copyright, and existing IP law, setting aside the massive energy costs and concomitant resource scarcity and environmental impacts, AI was not a tool created to unlock the creative potential of millions. No one is spending a trillion+ dollars of capex so you have an easier time writing an article.
Companies are spending a trillion+ dollars to not need expensive workers. AI exists to give capital access to skill, while denying skill access to capital. For all your examples of technologies superseding more laborious processes, you ignore the fact that they all pretty much *improve* on the former labor. I don't have to worry about my sloppy arithmetic mistakes when I use a calculator that is much faster and more accurate. My dishwasher can tolerate water far hotter than my hands can, enabling a far more sterile clean than I can achieve without it (and it uses far less water than I do when washing by hand). The skill I've given up is significantly outweighed by the virtues of the new substitute.
That is not true with AI. Coders are losing skill, and losing the experiential understanding of the code they're still responsible for. And the flood of insecure AI slop apps belies any argument that what we're giving up is worth what we're getting in return. And as far as writing? AI continues to beclown writers putting trust into the output of AI. If accuracy and trust mean anything to the writer or the reader, generative AI cannot be part of the process.
One can critically think. Or one can use statistically configured, lowest common denominator pablum as a crutch. Good luck, happy dogs.
I noticed a problem with one sentence: "Why is your concern." - shouldn't there be a "?" at the end of it? Given the subject of the article, I'd hope for "perfect" proof-reading of its contents.
No question mark, but arguably I could have written it like:
"Why" is your concern.
No, there shouldn't be a “?" at the end of that sentence. It's a perfectly valid shorthand for “(The reason) why is your concern.”
I had never considered this argument. Thanks for writing this article. I’ll think on it!
And I just noticed the Substack “Scan for AI” button. Ha!