Ryytikki
banner
ryytikki.bsky.social
Ryytikki
@ryytikki.bsky.social
small gay creature
November 16, 2025 at 12:00 AM
for an actual answer: any Brother laser printer and glock 19 to keep within visible range of it so it knows what will happen if it doesnt play nice
November 15, 2025 at 11:43 PM
the most reliable option is a small sickly victorian british boy with a can-do attitude but you can substitute with a wizened clergyman in a pinch
November 15, 2025 at 11:40 PM
our brains simply arent wired to handle things that arent human but can communicate human-ly
November 15, 2025 at 4:01 PM
your post is proof that its hard to even criticize LLMs without anthropomorphizing

lying implies an active world model and an intent to deceive. AIs that dont appear human are simply culled before they can be released to the world, the same way that animals who dont eat cant live/reproduce
November 15, 2025 at 4:01 PM
and at the same time trying to build an economy so reliant on it that any attempts to reign it in are met with fierce resistance by companies who use it
November 15, 2025 at 3:28 PM
I'd rather he didnt
November 12, 2025 at 4:39 PM
its spelt naturist
November 12, 2025 at 3:55 PM
and if the attacker is clever enough to make their spam look like legit traffic (spoofing FFXIV packets, in this case) its even harder to spot :(
November 11, 2025 at 12:42 AM
DDoS protections can only ever reduce the impact of DDoS attacks, not fully prevent them

Its especially bad with modern botnets, you can get tens of thousands of IoT devices (smart fridges, routers, doorbells, thermostats, etc) in a single attack if the attacker is dedicated enough
November 11, 2025 at 12:40 AM
1000% agree, my comparison vs cloth/hair physics is that imo its in the same "algorithmically solving something with an insanely huge phase space" area

The models, rigs, textures, etc are still made by humans, this just lets them move naturally no matter how complex of a world said humans build
November 10, 2025 at 10:01 PM
you'd be surprised by how much work goes into cloth physics models, i'm still seeing regular breakthrough papers (especially with stuff like twisting cloth)

either way, just because genAI uses machine learning, that doesnt mean that all machine learning is bad
November 10, 2025 at 9:36 PM
(meant genAI based things, not just LLMs, toddler induced sleep deprivation is frying my brain)
November 10, 2025 at 9:29 PM
adaptive/procedural movement is in the same category as stuff like physics-based hair and clothing animation. A useful tool to allow you to have complex motion without needing to put in an absurd amount of work
November 10, 2025 at 9:28 PM
yes, and this has basically 0 of the negatives of LLM based things (insane power/water usage, stolen training data, etc)

This technique has been used for adaptive animations for years. It lets you have movement that fits the world better without an insane and unfeasable amount of tedious work
November 10, 2025 at 9:21 PM
those wizards are gonna be pissed when they get to floor 8
November 10, 2025 at 4:18 AM
the only thing i remember is that theres an episode about an alien that fucks people so hard they disintegrate
November 10, 2025 at 4:08 AM
"kind of annoying small thingie that lives in the world and propagates" is what i aspire to
November 10, 2025 at 12:34 AM
my condolences that i had to :(
November 8, 2025 at 10:10 PM
shame this is AI :(
November 8, 2025 at 9:58 PM
nah the earth orbits at about 30km a second so if you travel any further than about 7 minutes into the past that wouldnt be an issue
November 8, 2025 at 6:04 PM
that not can bee that can moth
November 8, 2025 at 5:50 PM
not really, right wing populism is a growing issue in all of those countries

bigotry with social safetynets/healthcare is still bigotry, unfortunately
November 8, 2025 at 3:55 PM
and renner is already known for doing weird and unnerving stuff like whatever the fuck his app was so its not like he has some kind of pristine reputation that people dont want to immediately damage
November 8, 2025 at 1:12 PM