There’s this thing they do in mathematics that involves extrapolating something ad absurdum as a method of proof. You might be familiar with induction or its much better named sibling: proof by exhaustion.
I’ve never been able to get away from this line of thinking when evaluating a system. The system here is capitalism, and its reward function is money, which is transferable into capital, which can then be rented, thus earning money, ad absurdum.
Extrapolating, there is only one conclusion: a single entity ends up “owning” everything.
The system, of course, will collapse long before that, but that is the trajectory. Indeed, I would argue that we are already in the process of collapse.
Except the respect of other people. Billions of them. All minus one. Hungry and frustrated. Who could this entity trust? Being the most valuable target in the world doesn’t sound very safe.
Rather than “Artificial Intelligence” it is more like “Artificial Stupidity.”
I have found it useful in editing something I have created provided that I carefully defined the editing prompt. But, when I tried other things I was disappointed in the extreme.
I love this story. As I understand it, he also invented the concept of the 5-day work week… again not out of kindness, but because unless people have weekend leisure time they don’t need cars to go anywhere!
I’m choosing to interpret this as “I have absolutely no desire to live in a world where pigfender is right.”
Aha, you’re thinking like a rational, ethical human being who has empathy and logic.
They don’t care about the future. They care about maximising personal investment returns. This is the same reason why government legislation is required to stop companies polluting, etc etc.
But… some good news to end on:
Erm. Hang on…
Yeah, I’ll be there in a minu…
Okay, yeah, found some: 4 months ago, a child was reunited with their lost toy elephant.
I’ve put a like to the above, although in my case as long as it is not forced on me (as long as it can be turned off), I’m fine.
I have the feeling it is not so much gonna happen, anyways.
(I run the Windows version, for what it matters. Saying in case it is actually the OS that finds its way to mess up (I’ve redacted the F word, here) what is none of its business.)
The staggering salaries are in part because the number of people in the world with PhDs in relevant fields is relatively small, and several companies with effectively unlimited resources are in a bidding war to hire them.
Those aren’t the “people driving this,” though. The drivers are the CEOs and affiliated marketers who are making ever more expansive claims in order to compete for VC money. Even if the actual researchers know how unrealistic those promises are, they aren’t going to turn down such huge piles of money.
Oppenheimer worked for the government. That’s never paid as well as equivalent expertise in the private sector.
It is effectively in every app now, due to the integration that the OS performs. Thankfully, you can turn off Apple’s Language Tools, but I’m not sure about Windows.
Copilot doesn’t offer any particular tools in menu options for Scrivener the way Apple Intelligence does on MacOS.
The new breed of Copilot+ PCs in the Windows world are, in the insider releases, dropping tools that can interact with any form of text, whether true text or text as part of a picture, but nothing like Copilot, which I suppose, can analyse a written piece if cut from Scrivener and pasted into it.
But Copilot itself does not attach itself to Scrivener in any way.
There are various ways for Windows users to turn off (disable) Copilot from Settings, to advanced methods like Group Policies and the Windows Registry.
It’s like all other nuisances junk no one asked for in the first place, like widgets curated for you by who knows who and shoved down your throat and useless info feeds geared at creating generation numbness.
The next mess on the horizon is an expanded Start Menu of every app you never wanted in widgets of folder and sub-folders, right down to rubbish like ways to wash Aunt Pitypat’s pantaloons.
I think someone else here nailed it, the only reason I’m interested in really using Scrivener at this point is that it doesn’t have AI integration. AI integration would provide a mechanism for the models to essentially be trained on what you’re writing. Training is really just another word for stealing, both the style of your writing and potentially even your ideas. People don’t seem to realize that AI is just regurgitating what it has been trained on. Integration with AI is a really really bad idea.
The issue is that the OS you’re using has AI features, which means every app you use passively gets these features. For example, if you have the latest macOS, I believe the language features are now enabled by default (if not yet, they will be soon), meaning every single app with text input (including Scrivener) will get those AI features. It’s not Scrivener/your app doing it, it’s the OS. I imagine this will be the norm for newer versions of Windows, too. These are things for you to disable, but I think most folks won’t disable them, so it will very quickly become the norm to have AI assistance in anything you do.
(I hate this on many levels, but that’s not the point of this response.)
This week’s comments by David Weston, Corporate Vice President of Enterprise and OS Security at Microsoft about the future of Windows, is causing waves—waves of ridicule with statements on mouse and keyboard use being foreign tools and Agentic AI integration as the be all and end all of everything.
I wonder if the 2030 vision is fuelled by the much needed contextual topic search function, i.e. Agentic AI deployment in Windows 11 Settings App (a tool for users to destroy their machine’s capabilities without much effort that no one asked for, as a replacement for an easily understandable Control Panel). Of course, you need a Copilot+ PC for Agentic AI on Windows.
With visions like this, Scrivener 4 for Windows—well, would it be a viable platform for the app? Will L&L Customer Support offer the most wonderfully delusional answers with Agentic AI?
I used ChatGTP to proof-read my novel. I let it focus on spelling, grammar and consistency. For the most part it worked really well.
However, I had to copy & paste the text into ChatGPT and then transfer its corrections back to the text. So now I’m wondering if there are any plans to integrate AI into Scrivener in a way that it will suggest such corrections inline.
BTW, I tried the built-in Apple Intelligence proof-reading but it will just suggest the whole new text without the corrections highlighted, so it’s pretty useless.
It’s been mentioned elsewhere by the devs and L&L staff that Scrivener will not incorporate any AI functions beyond its ability to access the OS’s AI tools and any third-party AI-enabled software that can integrate with Scrivener.
(If I’m mistaken then a correction will be forthcoming by said staff.)
