106 points by atugard about 22 hours ago | 87 comments | View on ycombinator
FinnLobsien about 21 hours ago |
barbariangrunge about 19 hours ago |
Wont the safety systems of your providers automatically save all this history indefinitely and then it will be out there forever to be used against us in some way? Eg, suppose a custody hearing or professional conduct review or a trial or even a job interview or insurance claim?
Especially if its reading and researching in the background as you go
dgrcode about 20 hours ago |
We need more tools of this kind.
HaloZero about 19 hours ago |
My wife is a writer and her and her community go to great lengths to avoid having their writing being used to train models. It's a privacy thing but it's also I don't want my ideas thing.
yakkomajuri about 19 hours ago |
It's helped me with design [1] but I've also been building tools in this vein as experiments.
mizzao about 12 hours ago |
Looking forward to giving this a try!
JoeOfTexas about 19 hours ago |
I went to great lengths to improve the precision of scans across an entire book. You can read about it in my blog here: https://authrot.app/blog/how-authrot-reads-a-novel
liamnorm about 15 hours ago |
Much have I traveled in the realms of gold
And many goodly states and kingdoms seen;
Round many western islands have I been
Which bards in fealty to Apollo hold.
Oft of one wide expanse had I been told
That deep-browed Homer ruled as his demesne;
Yet never did I breathe its pure serene
Till I heard Chapman speak out loud and bold:
Then felt I like some watcher of the skies
When a new planet swims into his ken;
Or like stout Cortez when with eagle eyes
He stared at the Pacific—and all his men
Looked at each other with a wild surmise—
Silent, upon a peak in Darien.klntsky about 18 hours ago |
I'm working on something very similar: https://x.com/klntsky/status/2123140693954900007
With AI reviews of prose, I noticed that you really have to keep separate nitpicking rules per document for them to be useful, so I focused my solution on that exactly.
You create a "reader" (an agent), and whenever an issue is flagged you are offered multiple way to change the rules to allow it, or you can set your own rules. The agent NEVER edits or proposes you text. You can edit manually and the flagged issue disappears.
Not sure if I will publish it though - that loop needs hours of dogfooding and I don't write much.
Feel free to DM if you want to test it
vr46 about 18 hours ago |
Gonna try Foolscap
YuriyChobotok about 19 hours ago |
iamben about 19 hours ago |
iamcalledrob about 20 hours ago |
I think we have jumped to the "switch brain off" style of AI use (unsurprisingly because it's the path of lease resistance and of most immediate progress). I really hope there's an alternative world where the technology can be used to engage your brain more than ever.
bennyg about 19 hours ago |
I'm working on another children's book right now, so this is right up my alley!
janandonly about 20 hours ago |
Is this tool any good for these kinds of text?
happythenewsad about 15 hours ago |
I'm trying to solve the same problem from a different angle - increase the output and quality of human writing (specifically, fiction).
It's soft-launched and early days; would love some suggestions on what features to add:
penmightier.com
richwalkup about 18 hours ago |
JoeAltmaier about 16 hours ago |
chmod7000 about 19 hours ago |
guy4261 about 20 hours ago |
[0] https://www.cbs17.com/news/ap-prosecutor-alleges-ex-nfl-play...
undefined about 19 hours ago |
acrophiliac about 18 hours ago |
acrophiliac about 18 hours ago |
bdashdash about 18 hours ago |
Could this be extrapolated to learning in general?
LLMs are great as a personal tutor, but it's tempting to just let the AI do the thing for you.
Your approach to restriction might have further applications.
Something to think about.
chiasabah about 19 hours ago |
djeastm about 20 hours ago |
HaloZero about 20 hours ago |
0gs about 21 hours ago |
dleeftink about 21 hours ago |
The pentimento feature ('git for prose') is likely something we'll start seeing more in other authoring environments going ahead.
iknownthing about 20 hours ago |
noir_lord about 19 hours ago |
It's also largely the way I use AI currently for development, I write the code, then I throw it at the clanker to review, it comes back, I read what it output carefully, if I agree I alter my code, if I don't I leave it out (and update errata.md to say why for the future if I strongly disagree).
It's essentially treat as a slightly smarter rubber duck but the code is still mine and I still understand what it does vs just throwing it over the fence and running.
bwilli123 about 18 hours ago |
dond1986 about 20 hours ago |
undefined about 22 hours ago |
ghazanfarabas61 about 1 hour ago |
nortiztech about 21 hours ago |
Now that we're realizing the writing, coding, designing process is actually important and many of us enjoy them, these kinds of tools are great.
AI tools that work with you and augment your process, but don't replace it.