Gruff
fropome.bsky.social
Gruff
@fropome.bsky.social
Damned fool
I love the snobbiness of 'it was meant to be spoken', coupled with the I'm-so-smart 1.25x speed. Authenticity is important, but my mind just works too fast bro.
November 11, 2025 at 9:36 AM
As an old-time webcomic fan, it used to be that I had all my faves in my favourite bar and visited them when I knew they'd update. With new artists, I tend to just see their latest strip on reddit. Def harder for them imo.
October 21, 2025 at 5:01 PM
Yes well done, I am too upset by my thrashing to participate further. Have 1000 Internet points.
October 17, 2025 at 3:01 PM
I'm really not an evangelist for LLMs. I think they're quite limited and vastly oversold, but they do have some uses too. I don't care enough about them to continue this further sorry.
October 17, 2025 at 2:54 PM
Well no. I have a 90 minute meeting. The AI spits out a summary of bullet points and actions. I read it - maybe make a quick change, maybe not - and it's done. For me this is quicker than writing the summary myself, and means I can concentrate more during the meeting.
October 17, 2025 at 2:52 PM
In the sense that a summarised meeting transcript is data, yes. You win. Well done.
October 17, 2025 at 2:36 PM
I have said I don't use it for data gathering. In meeting software it'll automatically produce notes afterwards... While I still need to check them, they're mostly fine and quicker than starting from a blank page. In code debugging the AI will sometimes spot a silly mistake I can't see.
October 17, 2025 at 2:30 PM
I don't care if it does that, because I only use it as a first draft and correct it.
I've used LLMs on and off for a couple of years now, and imo they're fine but limited. They're also vastly oversold, artists should win royalties from them, and I'm not sure they'll ever make money.
October 17, 2025 at 2:28 PM
I said that LLMs can draft OK text based on set information where style isn't important, but should still be checked, and shouldn't be trusted for information gathering. That makes me seem like someone who wants to marry AI?
October 17, 2025 at 2:08 PM
I don't use ai for analysis. I use it for first drafts of meeting notes and similar, and sometimes it helps to debug code.
October 17, 2025 at 1:55 PM
Data analyst. I don't actually use LLMs much, and am not sure they're cost effective. I do think they have some uses as long as you're careful.
October 17, 2025 at 12:56 PM
OK
October 17, 2025 at 12:44 PM
Like I said originally, drafting only. I trust it less than I would trust (say) a brand new staff member. Doesn't mean they're useless.
October 17, 2025 at 12:33 PM
You may think so, I don't. When I do creative writing or personal emails, yes. If I'm trying to write minutes, no.
October 17, 2025 at 12:16 PM
If you're writing a novel or a YouTube essay, sure. If you're just writing a summary of a meeting or desk notes for a process, that's not true.
October 17, 2025 at 12:00 PM
I think LLMs are good at first drafts or revisions, where your personal tone isn't important. I wouldn't use them for information gathering. I'm not convinced they're transformational or are a good product per cost, but they are useful.
October 17, 2025 at 11:29 AM
As an interested idiot, it feels like Brexit / Johnson meant the Tories had to at least pretend to be idiots for so long that by the time they lost the election it was too much of a habit. Only those capable of thriving in such an environment were(/are) in leading positions.
October 7, 2025 at 11:59 AM
French govt not being bothered by whether the English is correct doesn't seem entirely off-brand? Us anglophones should have learned to French.
October 7, 2025 at 9:49 AM
Yeah agreed - it wouldn't be *wrong* here, but it's why I got confused. Thanks!
October 5, 2025 at 7:32 PM
Oh I see - so by govt you mean the entire executive, and by regime you mean the leadership?
I think this is a context thing, in a UK context I don't think people would use those words the same way (most of the non-elected parts of govt are impartial, so "govt" *usually* means the elected bit).
October 5, 2025 at 7:28 PM
Still not sure I follow!
I see that there's a difference between a government which wants to be fascist and one which manages that - if that's what you mean?
October 5, 2025 at 7:19 PM
Genuine question - what do you mean by the difference between government and regime here?
October 5, 2025 at 7:05 PM
I think that's a really interesting idea... Tho personally I don't feel like communicating with chatgpt/ Alexa is the same as a conversation. It feels more like dictating a line of code - my focus is on having clear syntax. Perhaps like telling a dog to sit.
October 5, 2025 at 12:35 PM
I know you see him as threatening of course. I'm surprised that you think there are legal limits for him - I'm interested in your thoughts not trying to be argumentative. You're better informed than me I'm sure!
Do you think there's no risk that the US becomes an autocracy in the next 5 years?
September 11, 2025 at 12:05 PM