Forum: Story Discussion and Feedback

Not really, some just appreciate using it, while others—like me—are simply old-school, not needing helping hand to write. Well, I do, but I prefer a helping hand with some knowledge of what they're doing.

It all depends on HOW you're using AI. As a research assistant and an analysis tool (e.g., building a story bible and looking for plot holes, continuity errors, and anachronisms), it's great, if you learn how to use it effectively.

It's a tool in my toolbox, like my text editor and publication tools. The publication tool had a learning curve, just as AI does. Just as MS Word (which you use) does.

The fact that others use different tools doesn't make them any less a craftsman than you claim to be.

My prose is 100% mine, but using AI as noted above does work that no human could reasonably do unless I paid them more than I earn from my writing (which is more than enough to support myself).

It all depends on HOW you're using AI. As a research assistant and an analysis tool (e.g., building a story bible and looking for plot holes, continuity errors, and anachronisms), it's great, if you learn how to use it effectively.

It's a tool in my toolbox, like my text editor and publication tools. The publication tool had a learning curve, just as AI does. Just as MS Word (which you use) does.

The fact that others use different tools doesn't make them any less a craftsman than you claim to be.

My prose is 100% mine, but using AI as noted above does work that no human could reasonably do unless I paid them more than I earn from my writing (which is more than enough to support myself).

This. All of this.

AI is a tool. Some people misuse new tools. They always have. They always will.

The issues isn't the tools, it's the rampant abuse of them, as in the write hands, they are beneficial, yet those mostly relying on them, they're again, mostly using them as a crutch, never learning to walk on their own.

So I'm mostly bemoaning the fact that they're not advancing, as writers, and they're entirely too dependent on the LLM AI to co-write the story for them.

Again, it's all about known your tools, and using them as they were designed for, rather than hammering nails with a precision screw driver, a drill bit or a baseball bat.

As at this point, I'm again fewer stories than I ever have on SOL, which is not a highlight of the site. They're doing well, but … the overall writing isn't what it once was.

Generally, from what I've seen, those relying on LLM AI were simply not adept at writing fiction, as those AIs gave them a way in, yet because they're relying on AI, they really aren't learning how to write.

And in the end, that is the biggest issue I have with it.

Writing is like almost anything else, from learning to paint to playing an instrument and driving a vehicle. One can only improve in ability by actually doing it. If you never do it, you will never improve.

Now for myself, I have been writing "erotica" for over three decades. And I can look back at my older stories and cringe, but that is also what 90% of them were like in the era of BBS and Usenet. But over time I did improve.

Myself, I largely see it as a lazy crutch. And I bet that in 10 or 20 years it will have improved little, while those who actually do write will improve.

Based on the current rate of progress of LLMs, and assuming we don't hit either (on the positive side) actual AGI or (on the negative side) some currently unknown barrier to improved computational performance, memory capacity, and algorithm development, I would bet that we'll see considerable progress in 10 to 20 years.

In this case, the 'those who actually do write' are the LLM developers (hardware and software). Given the current rate of progress, guessing we've somehow plateaued right now strikes me as akin to the old joke that we do not need patent clerks because everything has already been invented.

I cannot help but wonder how long until I am sitting in front of a screen with an AI on the other side?

Yeah, it is bad enough for tales and artworks. But what comes after that?

Whatever it is, it will be ugly.

While the current LLM/AI systems aren't proficient, the way differential diagnosis works is literally algorithmic. List the symptoms, then proceed from most likely/easist to treat to rarest/most difficult to treat.

There are pocket diagnostic guides in use; this is just taking it to the next level.

The same is true for prescribing medication. Most physicains have to look up anything that isn't in the top prescribed drugs in Physicians Desk Reference.

Plugging the symptoms/diagnosis into an algorithm is simply automating how humans behave, and will improve over time.

This is not an argument for physicianless medicine, just giving the physicians additional tools.

the way differential diagnosis works

I believe it's defective and should be replaced by a game theory approach, taking factors like urgency and lethality into account.

Sepsis is very rare (many GPs never see a case in their careers) so is well down the priorities of differential diagnosis. But it's lethal and has to be treated as quickly as possible. So UK doctors get a sticking-plaster reminder, "Are you sure this isn't sepsis?" when doing their differential diagnosis. I believe "Are you sure this isn't cancer?" is on its way too - pro-patient activists recommend asking your GP that question anyway if there's any doubt, because it's another one often missed by differential diagnosis.

AJ

I believe it's defective and should be replaced by a game theory approach, taking factors like urgency and lethality into account.

Medicine changes slowly, and there is a rationale to it, but it's a lengthy conversation with significant nuances. You can read all about it in my series Good Medicine, starting in Book 5 and continuing through the (being written) Book 11.

Except, in medicine, it's not the standard treatments which are the problems, it's the exceptions, when someone's body doesn't respond as expected.

Individuals are not cogs in a machine, they're often unique, so a 'one-size fits all' approach only helps certain individuals, not everyone.

Since one idjit put me on Lithium, completely ignoring the warnings at the time, I've had a lifelong host of medical issues (i.e. my body typically reject most treatment, thinking they're attacking than than helping my body).

So for all of you without background, thank your lucky stars, but it's been a literal nightmare for me. One which could easily have been avoided.

Except, in medicine, it's not the standard treatments which are the problems, it's the exceptions, when someone's body doesn't respond as expected.

Individuals are not cogs in a machine, they're often unique, so a 'one-size fits all' approach only helps certain individuals, not everyone.

While this is true, it's also the case that any system that delivers mass services has to be geared towards helping the vast majority and achieving the best overall success. There is truly no other cost-effective option, and resources aren't infinite.

My solution to my unique set of medical concerns is research and conversations with my physician. I actually know more about my conditions than he does, and this is not uncommon, because GPs are, well, general practitioners.

So, I do my research, discuss it at length with him, and, when appropriate, seek help via specialists.

Why? Because I can't simply go to the pharmacist, explain what I need, and obtain it. I have to go through the medical system, with socially-agreed gatekeeping.

Medicine can only work if patients are actively participating in their own care. The fundamental problem with 'modern medicine' is that they do not.

If you are the outlier, you have to be proactive. A mass-care system cannot provide what you need without that. Right or wrong, it's the truth, so long as there are limited resources (time, money, qualified physicians, qualified nursing staff, etc.)

What bothers me is those human physicians that cannot or will not take the time to answer the difficult questions. I ran into a specialist a few years ago that gave me the line, "Because I said so". Needless to say, I did not go back and found another that would at least attempt to answer my concerns

That is a red flag if there ever was one.

My most lengthy debates have been around statins, where my GP was pushing the 'party line' about 'risk' which is relative risk, not absolute risk.

If the study says that without the drug, 6 out of 10,000 had heart attacks, and with it, only 3 out of 10,000 had heart attacks, they claim a '50% reduction', which is 'true' as far as it goes, but the reduction is better reported as 0.03% (i.e., 3/10,000) so you can understand the true risk/reward.

Given the side effects of statins, and given I had never had a coronary event, a 0.03% reduction in heart attack risk was simply not worth it.

As always, we need to be our own medical advocate, as who else will be as focused on our health than we are, and who better understands our health than us.

People with normal problems and conditions can largely skate on by, but those of us with either unusual, non-standard or complicated conditions pretty much have to take charge.

That said, the reason wry I stopped writing is because my psychiatrist and I got into a tiff over my care—insurance companies began issuing strict limits on 'time allotted for care', when I'd pretty much taught her everything she knows about my Spectrum-Disorder condition.

So despite knowing that the medication needs to be boosted after a time, due to a natural 'immunity' to the standard dosages, they refused and I simply stopped going.

Now that I'm at the point where my concentration is SO bad that I no longer remember to take it, I've learned something new.

I still can't remember much of anything, yet suddenly, my physical problems (troubles standing, balancing, etc.) have now largely vanished, meaning I can now take a shower without any risk of falling (a serious concern for well over a dozen or so years).

So the trade off calculation is now entirely different. So, do I really need to write, if I can instead focus on other activities (reading, returning to my long-abandoned artwork, etc.), I can probably do much better never writing again (other than in posts, like this one).

Otherwise, I may try learning to write short-stories, short enough that I don't need to worry about protracted thirty-chapter story arcs. Heck, long hikes may even be in the cards again!

I still can't hold things anymore (a separate issue) as I have little motor control on my predominate (right) side, to the point that medicine (shots, like daily insulin) are only partially effective on that side of my body, as are blood tests (20 to 35% differences in results).

Still, that's a reasonable trade off for what I'll like never be able to do again, no matter what I do.

But as far as memory is concerned, I can barely remember what happened in the stories I read on a monthly basis, or even who most of the characters are after so long. :(

In the end, it's usually one condition or the other, as at this age, everything starts degrading, so best to stick with what's working than stressing over what won't.

In the end, it's usually one condition or the other, as at this age, everything starts degrading, so best to stick with what's working than stressing over what won't.

Exactly. That was the most recent conversation I had with my GP. I laid out my exact priorities and made it clear I would not sacrifice certain things, even if it increased the risks in other areas. So far, I haven't had to make any significant tradeoffs, but I know the day is coming.

That said, my dad lived to 102+ and his health was excellent until the final year of his life, so I have that going for me, which is nice. (Obl. Darmok)

I think the subject of AI now deserves its own Forum topic, on a par with Author Hangout etc.

AJ