Axe Ghost. Now with Fragments mode
@axeghostgame.bsky.social
Cleverly push monsters around a grid and blow them up.
Demo👉 https://store.steampowered.com/app/2994110/
Demo👉 https://store.steampowered.com/app/2994110/
whatever the downsides, if you *successfully* used AI for everything, in an academic context (e.g. synthesizing novel research), that would involve learning v relevant and valuable skills
January 23, 2026 at 7:46 AM
whatever the downsides, if you *successfully* used AI for everything, in an academic context (e.g. synthesizing novel research), that would involve learning v relevant and valuable skills
“Did you get that new 'pull' icon done for the doors?”
“Sure did, boss. Shows a real terrified hopeless struggle just like you wanted.”
“Sure did, boss. Shows a real terrified hopeless struggle just like you wanted.”
January 20, 2026 at 5:31 PM
“Did you get that new 'pull' icon done for the doors?”
“Sure did, boss. Shows a real terrified hopeless struggle just like you wanted.”
“Sure did, boss. Shows a real terrified hopeless struggle just like you wanted.”
that claim is directly supported in the next two paragraphs.
January 19, 2026 at 1:20 PM
that claim is directly supported in the next two paragraphs.
huh, valve has changed the wording on the AI section of the content survey. which changes the meaning of 'yes'. were any devs who'd filled this out notified about this? setting the answer back to 'no'.
January 17, 2026 at 7:16 PM
huh, valve has changed the wording on the AI section of the content survey. which changes the meaning of 'yes'. were any devs who'd filled this out notified about this? setting the answer back to 'no'.
no. the article explicitly mentions ~650–700 gallons as a commonly cited public figure *that it then rejects for its own analysis*, and explains why.
January 17, 2026 at 7:44 AM
no. the article explicitly mentions ~650–700 gallons as a commonly cited public figure *that it then rejects for its own analysis*, and explains why.
current claude workflow for godot project
* write step doc (.md) describing a feature + acceptance tests
* using step doc, py script loads relevant files, runs GDUnit tests, builds minimal context
* claude implements feature, writes tests
* tests fail? invoke claude w error log until they don't
* write step doc (.md) describing a feature + acceptance tests
* using step doc, py script loads relevant files, runs GDUnit tests, builds minimal context
* claude implements feature, writes tests
* tests fail? invoke claude w error log until they don't
January 16, 2026 at 3:36 PM
current claude workflow for godot project
* write step doc (.md) describing a feature + acceptance tests
* using step doc, py script loads relevant files, runs GDUnit tests, builds minimal context
* claude implements feature, writes tests
* tests fail? invoke claude w error log until they don't
* write step doc (.md) describing a feature + acceptance tests
* using step doc, py script loads relevant files, runs GDUnit tests, builds minimal context
* claude implements feature, writes tests
* tests fail? invoke claude w error log until they don't
Post a game you remember playing that nobody else remembers, and which also rules
January 15, 2026 at 3:01 PM
Post a game you remember playing that nobody else remembers, and which also rules
most heavy users of llms - people who find this tech very useful daily - are far from dumb ime. it's a very interesting group.
January 15, 2026 at 2:01 PM
most heavy users of llms - people who find this tech very useful daily - are far from dumb ime. it's a very interesting group.
this site already uses AI in content moderation and recommendations
January 15, 2026 at 6:29 AM
this site already uses AI in content moderation and recommendations
suggestion: read the linked post you're commenting on to learn what they're trying to accomplish (it's AI used in the areas it's *already being used* for this site)
January 15, 2026 at 6:19 AM
suggestion: read the linked post you're commenting on to learn what they're trying to accomplish (it's AI used in the areas it's *already being used* for this site)
it's been here all along. for his kind of purpose:
January 15, 2026 at 6:16 AM
it's been here all along. for his kind of purpose:
you don't want these things?
January 15, 2026 at 6:14 AM
you don't want these things?
ohh this is cunning. gwern.net/blog/2025/be...
January 14, 2026 at 8:03 PM
ohh this is cunning. gwern.net/blog/2025/be...
parking the new thing. was fun to work through! learned a bunch. it fell into slay the spire's gravity well really hard; the problems in the near design space have been solved so well in that game. felt like a choice between 'be too derivative' or 'do something clearly worse', neither feels good :D
January 13, 2026 at 7:27 PM
parking the new thing. was fun to work through! learned a bunch. it fell into slay the spire's gravity well really hard; the problems in the near design space have been solved so well in that game. felt like a choice between 'be too derivative' or 'do something clearly worse', neither feels good :D
'AI is a tool for firing people' makes exactly the same amount of sense, as a summary, as:
'cameras are a tool for making painters destitute'
'the combustion engine is a tool for demolishing the equine economy'
we all see that right?
'cameras are a tool for making painters destitute'
'the combustion engine is a tool for demolishing the equine economy'
we all see that right?
January 13, 2026 at 10:29 AM
'AI is a tool for firing people' makes exactly the same amount of sense, as a summary, as:
'cameras are a tool for making painters destitute'
'the combustion engine is a tool for demolishing the equine economy'
we all see that right?
'cameras are a tool for making painters destitute'
'the combustion engine is a tool for demolishing the equine economy'
we all see that right?
if prompt-based processes cannot produce "human connection" then neither can handmade art.
this is the same magical thinking that denied photography and readymades legitimacy. we've had more than 100 years to get used to this idea😂
this is the same magical thinking that denied photography and readymades legitimacy. we've had more than 100 years to get used to this idea😂
January 12, 2026 at 1:31 PM
if prompt-based processes cannot produce "human connection" then neither can handmade art.
this is the same magical thinking that denied photography and readymades legitimacy. we've had more than 100 years to get used to this idea😂
this is the same magical thinking that denied photography and readymades legitimacy. we've had more than 100 years to get used to this idea😂
one weird trick!
January 12, 2026 at 9:55 AM
one weird trick!
skeptics still exist who insist AI can't do "real math". meanwhile ChatGPT 5.2 & Aristotle are producing verifiable proofs on problems humans couldn't close for decades. what a time to be alive.
January 11, 2026 at 1:31 PM
skeptics still exist who insist AI can't do "real math". meanwhile ChatGPT 5.2 & Aristotle are producing verifiable proofs on problems humans couldn't close for decades. what a time to be alive.
it's sad that many have been mislead, and are still being mislead, that llm use is environmentally damaging (compared to other accepted activities)
bsky.app/profile/axeg...
bsky.app/profile/axeg...
January 9, 2026 at 10:57 AM
it's sad that many have been mislead, and are still being mislead, that llm use is environmentally damaging (compared to other accepted activities)
bsky.app/profile/axeg...
bsky.app/profile/axeg...
this is false. at worst, every case *in which a signature of AI use can be recognised in the final product* is a sign of lack of care. you don't notice the cases that you don't notice.
the 'AI signature' isn't a property of AI use, it's a property of unskilled/lazy AI use.
the 'AI signature' isn't a property of AI use, it's a property of unskilled/lazy AI use.
January 6, 2026 at 4:25 PM
this is false. at worst, every case *in which a signature of AI use can be recognised in the final product* is a sign of lack of care. you don't notice the cases that you don't notice.
the 'AI signature' isn't a property of AI use, it's a property of unskilled/lazy AI use.
the 'AI signature' isn't a property of AI use, it's a property of unskilled/lazy AI use.
yes. also, this was with GPT-4o, six months ago.
January 6, 2026 at 3:10 PM
yes. also, this was with GPT-4o, six months ago.
nidhogg is hilarious. nidhogg 2, very similar on paper, isn't at all.
lots of possible reasons. imo a big one is that everything about n2 is zany and grotesque looking. it insists very hard that it's a Funny Game. while n1 played it ~straight; nonchalant emergent slapstick.
lots of possible reasons. imo a big one is that everything about n2 is zany and grotesque looking. it insists very hard that it's a Funny Game. while n1 played it ~straight; nonchalant emergent slapstick.
January 5, 2026 at 3:59 PM
nidhogg is hilarious. nidhogg 2, very similar on paper, isn't at all.
lots of possible reasons. imo a big one is that everything about n2 is zany and grotesque looking. it insists very hard that it's a Funny Game. while n1 played it ~straight; nonchalant emergent slapstick.
lots of possible reasons. imo a big one is that everything about n2 is zany and grotesque looking. it insists very hard that it's a Funny Game. while n1 played it ~straight; nonchalant emergent slapstick.
approaching 200 hours on newthing dev. still not ruled it out, and still not validated it definitively 😅
December 29, 2025 at 12:21 PM
approaching 200 hours on newthing dev. still not ruled it out, and still not validated it definitively 😅