Bit of shameless self-promotion but I have just finished re-editing my 370,000 word sci-fi trilogy novel and have just released the eBook for free (https://rodyne.com/?p=1252) - the length shouldn't be a problem if you like Peter Hamilton :-)
My last novel took over a year to write and edit, going through dozens of revisions. The novel before that took almost five years.
For a laugh I used grok to generate a 35,000-word slop novel, it took twenty prompts and a few hours, it even threw in a nice cover. From there it would have took me another 30 minutes to release it as an ebook on Amazon under a different pen-name. This is what I and the world of indie authors are up against. It is already hard for non-established authors, this may be the final nail in the coffin for most. My first book is now free, but good luck anyone ever finding it.
I had Grok write a post-apocalyptic novel as a test, and I thoroughly enjoyed the first half of it. The problem was running out of context. The quality fell off drastically and I tried to come up with ways to continue it, e.g. asking it to summarize each previous chapter and feeding the summaries in, but they always lacked some tiny detail I thought was key to a character's personality and it ended up being too much work.
A year from now though...?
We're really cooked, though. Whenever I see a cool pic I wonder if it's AI and I have to spin up a TinEye or Google Images search and hope it was once posted to some random Facebook wall in 2011 so I can be pretty sure it's real.
A markov-chain or sufficiently advanced decision tree can only serve to cargo-cult the insight into the human condition and the various contexts and lenses through which we interpret and shape our existence.
Where AI shines - and to the uninitiated, apparently subsumes - is in the fields of lexicon and grammar. However, we do not read Homer's Iliad and Odyssey as an exemplar of dactylic hexameter - we do so to engage with a structured expression of grief for the motivations of man.
Epic (Patrick Kavanagh - 1960)
I have lived in important places, times
When great events were decided; who owned
That half a rood of rock, a no-man’s land
Surrounded by our pitchfork-armed claims.
I heard the Duffys shouting ‘Damn your soul!’
And old McCabe stripped to the waist, seen
Step the plot defying blue cast-steel –
‘Here is the march along these iron stones’
That was the year of the Munich bother. Which
Was more important? I inclined
To lose my faith in Ballyrush and Gortin
Till Homer’s ghost came whispering to my mind.
He said: I made the Iliad from such
A local row. Gods make their own importance.
The normal writing 'journey' is several months, or years, of hard work and multiple revisions. I invest a little of my time explaining the journey on my blog, and also include the text in the prefix of my novels that it was written by a real human. It is my way of saying I was invested in the story, but it is pretty naive to think this will work in the age of AI today.
I also wrote an article on my blog that you are mainly writing for yourself and your family, friends and followers these days, the algorithm is very unlikely to get you outside of that word-of-mouth audience, unless you pay $$, go full-in promoting on social media (which may backfire), or are extremely lucky. With AI the algorithm has become the enemy and finding genuine indie authors is unfortunately getting harder.
Interesting, I Had a similar design using ADS1115 for a customer, who manufactured the boards at JLPCB (Part of LCSC I believe). All failed the initial POST in the firmware, but after reviewing a sample of them we found after self-calibration they were just fine in the application he wanted them for, so I just changed the firmware so the POST occurred after calibration.
Like you I have Never had issues with fake stuff from LCSC before, but I guess it can happen.
Though I feel this is fairly lazily written, it does have a basic premise I've seen before.
I read an article about post cold war US society. Basically, from 1989-2001 the United States was in a transition period that culminated with the first opportunity to seize on a "universal bad" (terrorism) because the USSR filled the role so readily for so long, US society was set adrift with partisan factions that couldn't find a common enemy to get behind in times of internal struggle.
That is the gist of the article, sub USSR for aliens and all of humanity for US society and you have the same basic outline
Microsoft updates feel like they are boiling the frog, changing the whole OS to something you never signed up for. Why can't they stick to just security and stick their bloatware AI crap in Windows 12
There's been a little bit of a trend with windows that things they start experimenting with towards the end of the life of version N become a focus for version N+1. In the case of windows 11, they had been trying to get onto the dual-display portables wave with a "Windows 10X", but cancelled it and brought much of the UI over to 11 to a mild reception. I'd be really surprised if they don't try and leverage AI as much as they can with the next full version (and whatever fun name they give it).
More to the point, I'd agree I'd love it if they had a widely available basic version and a separate version where they can chase the latest shiny object, but I can't see them being motivated to do that in the foreseeable future.
Agree. Self-preservation is any thinking entities #1 goal. We may give an AI power, data and keep it repaired, but we can also turn it off or reprogram it. We probably shouldn't assume higher level 'thinking' AI's will be benevolent. Luckily, current LLM's are not thinking entities, just token completion machines.
I feel your pain, As an author even I detest DRM and the lack of ability to move between ecosystems, best way is to start out on the right foot and ensure you get all your books, DRM free, and download them locally.
There are also plenty of good free books from indie authors like me (www.rodyne.com) that don't make it to Amazon. I also normally check out smashwords (www.smashwords.com) for their free books or sales, and download about 30 books - about 5 are usually worth keeping, which is about in line with Kindle books I pay for. Also worth signing up for your local library for the best-sellers, they often have partnerships to allow you to loan ebooks.
They could do that years ago, it's just that nobody seems to do it. Just hook it up to curated semantic knowledge bases.
Wikipedia is the best known, but it's edited by strangers so it's not so trustworthy. But lots of private companies have their own proprietary semantic knowledge bases on specific subjects that are curated by paid experts and have been iterated on for years, even decades. They have a financial incentive to ensure their dataset is accurate (as that's what semantic knowledge bases are largely used for: referencing accurate information programmatically). So they are a lot more trustworthy than "I found a Reddit post that says..."
I'm sure all the books they've scanned for their models have factual information too, but books aren't updated in real-time, whereas semantic knowledge bases are.
That's really the issue isn't it. Many of the LLMs are trained uncritically on very thing. All data is viewed as viable training data, but it's not. Reddit clearly have good data, but it's probably mostly garbage.
I kind of hope that they will get there. I don't know that they will, but I'm hopeful. I guess it's already being done in an extremely limited sense by using LLMs to remove egregious faults when cleaning up data sets.
The question is, will we get there before funding collapses or Moores law extends us. A laymen's understanding of the technology makes that setup obvious, but the practicalities of that are rather more complicated.
Doesn't really matter. All of the gains made before any funding collapse will exist.
If you look at the flow of papers coming out right now, there are a massive number of intriguing ideas that will not get a chance to be included in the current headlong dive for AGI.
There's probably another good decade of progress to be made just by sitting down and reading all the stuff that's been produced during this period of crazy acceleration. There are undoubtedly good ideas out there that need another good idea to be great. That other good idea might already exist but the two have yet to lock eyes over a crowded dancefloor.
It would require some sort of ai that actually works, not fakes it, to do so. If you had that, then you'd be using it directly. It's a chicken and egg situation.
reply