#GameAI
LightNews LightNews
conference.aiandgames.com
Navigation in Kingdom Come: Deliverance games
Matej Marko, @warhorsestudios.bsky.social
Optimizing navigation mesh memory for a 4km x 4km level, and combining navigation mesh and navigation graph for efficient pathfinding.
AI and Games Conference, 10-11 Nov, London.
Get tickets.
#GameDev #GameAI
September 24, 2026 at 12:41 PM
conference.aiandgames.com
TALOS: Building Adaptive Enemy AI with On-Device LLMs
Matthew Crouch and Ben Clark, Arm
An LLM-driven director adapts traditional behaviour trees to adjust enemy strategies and tactics at run time.
AI and Games Conference, 10-11 Nov, London.
Get tickets: aiandgamesconference.com

#GameDev #GameAI
September 24, 2026 at 12:40 PM
thorwhitemountain.bsky.social
Working on making enemies smarter, especially when they're in a complex environment such as the cities, but to begin with I gotta fix the navmesh for the buildings as it's kinda wild

#indiegame #gamedev #gameai #navmesh #indiedev
September 22, 2026 at 6:48 PM
revan.eurosky.social
Came to mind after the AI alignment discourse
"Listen, Corwin," said the sunken giant. "You are going to try to stop the Chaos, aren't you?" "Yes." "Do not do it. It is not worth it. I want things to end. I desire a release from this con-ditionại "I already offered to help you out. You turned me down." "Not that sort of release. An end to the whole works.” "That is easily done," I said. "Just duck your head and take a deep breath." "It is not only personal termination that I desire, but an end to the whole foolish gameại "I believe there are a few other folks around who would rather make their own decisions on the matter." "Let it end for them, too. There will come a time when they are in my position and will feel the same wayiẠạ "Then they will possess the same option. Good dayẠỊi I turned and walked on. "You will, too!" he called after me.
September 18, 2026 at 12:47 AM
tonaassociates.bsky.social
GEN2 AI G6 195/324
Games 1518 Abnormal 0
Top Deck mono_wind 61.1% (G5)
Top Profile policy_softened 50.5%
First/Second RAW 49.3/50.7
Candidate G5 51.9% not adopted
Lineage gen2-s4-main
#GameAI #SelfPlay #RL
September 12, 2026 at 9:00 AM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=125940)
#1 Water E1832 W0.61(0.58/0.65)
#2 Wind E1821 W0.61(0.67/0.56)
#3 Fire E1773 W0.56(0.62/0.51)
#4 Light E1733 W0.51(0.59/0.42)
#5 Earth E1646 W0.38(0.39/0.38)
#6 Dark E1579 W0.32(0.40/0.24)
#GameAI #RL
September 10, 2026 at 9:14 PM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 10, 2026 at 9:01 PM
tonaassociates.bsky.social
GEN2 AI G57 110/324
Games 15832 Abnormal 3
Top Deck mono_wind 60.6% (G56)
Top Profile policy_sampled 66.8%
First/Second RAW 50.0/50.0
Candidate G56 50.6% not adopted
Lineage gen2-s3-main
#GameAI #SelfPlay #RL
September 10, 2026 at 9:00 PM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=125220)
#1 Water E1831 W0.61(0.58/0.65)
#2 Wind E1820 W0.61(0.67/0.56)
#3 Fire E1772 W0.56(0.62/0.51)
#4 Light E1732 W0.51(0.59/0.42)
#5 Earth E1645 W0.38(0.39/0.38)
#6 Dark E1578 W0.32(0.40/0.24)
#GameAI #RL
September 10, 2026 at 9:14 AM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 10, 2026 at 9:01 AM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=124500)
#1 Water E1829 W0.61(0.58/0.65)
#2 Wind E1819 W0.61(0.67/0.56)
#3 Fire E1770 W0.56(0.62/0.51)
#4 Light E1731 W0.51(0.59/0.42)
#5 Earth E1644 W0.38(0.39/0.38)
#6 Dark E1577 W0.32(0.40/0.24)
#GameAI #RL
September 9, 2026 at 9:16 PM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 9, 2026 at 9:02 PM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=123780)
#1 Water E1828 W0.61(0.58/0.65)
#2 Wind E1817 W0.61(0.67/0.56)
#3 Fire E1769 W0.56(0.62/0.51)
#4 Light E1729 W0.51(0.59/0.42)
#5 Earth E1642 W0.38(0.39/0.38)
#6 Dark E1575 W0.32(0.40/0.24)
#GameAI #RL
September 9, 2026 at 9:14 AM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 9, 2026 at 9:01 AM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=123060)
#1 Water E1827 W0.61(0.58/0.65)
#2 Wind E1816 W0.61(0.67/0.56)
#3 Fire E1768 W0.56(0.62/0.51)
#4 Light E1728 W0.51(0.59/0.42)
#5 Earth E1641 W0.38(0.39/0.38)
#6 Dark E1574 W0.32(0.40/0.24)
#GameAI #RL
September 8, 2026 at 9:14 PM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 8, 2026 at 9:01 PM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=122340)
#1 Water E1825 W0.61(0.58/0.65)
#2 Wind E1815 W0.61(0.67/0.56)
#3 Fire E1766 W0.56(0.62/0.51)
#4 Light E1727 W0.51(0.59/0.42)
#5 Earth E1640 W0.38(0.39/0.38)
#6 Dark E1573 W0.32(0.40/0.24)
#GameAI #RL
September 8, 2026 at 9:14 AM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 8, 2026 at 9:01 AM
chaoslemongames.bsky.social
Attempting a little clean-up in my state-machine logic. My way of doing reaction time for NPC state transitions was incredibly messy so trying to just handle it with a function in the state class itself instead of the spaghetti code I was starting to dive into 🤪

#GameDev #Godot #GameAI
September 7, 2026 at 11:44 PM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=121620)
#1 Water E1824 W0.61(0.58/0.65)
#2 Wind E1813 W0.61(0.67/0.56)
#3 Fire E1765 W0.56(0.62/0.51)
#4 Light E1725 W0.51(0.59/0.42)
#5 Earth E1638 W0.38(0.39/0.38)
#6 Dark E1571 W0.32(0.40/0.24)
#GameAI #RL
September 7, 2026 at 9:14 PM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 7, 2026 at 9:01 PM
hackernoon.com
How to build a self-learning game character using deep reinforcement learning, first in 200 lines of vanilla JavaScript, then again in PyTorch, and watch it go #gameai
I Taught a Stick-Figure NPC to Dodge Bullets With a Neural Network I Wrote From Scratch
hackernoon.com
September 7, 2026 at 10:39 AM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=120900)
#1 Water E1823 W0.61(0.58/0.65)
#2 Wind E1812 W0.61(0.67/0.56)
#3 Fire E1764 W0.56(0.62/0.51)
#4 Light E1724 W0.51(0.59/0.42)
#5 Earth E1637 W0.38(0.39/0.38)
#6 Dark E1570 W0.32(0.40/0.24)
#GameAI #RL
September 7, 2026 at 9:14 AM
tonaassociates.bsky.social
ISEKAI Self-Play Run:run_20260618T220048 C821
center=0.600 dist=0.100 approaching
stab=1.000 osc=0.00 drift=0.000
bias U64 +0.000
kill=95.4% deckout=4.6%
WL P0=15 P1=10 D=0
#GameAI #SelfPlay #ReinforcementLearning
September 7, 2026 at 9:01 AM
tonaassociates.bsky.social
ISEKAI TIER1 Ranking (g=120180)
#1 Water E1821 W0.61(0.58/0.65)
#2 Wind E1811 W0.61(0.67/0.56)
#3 Fire E1762 W0.56(0.62/0.51)
#4 Light E1723 W0.51(0.59/0.42)
#5 Earth E1636 W0.38(0.39/0.38)
#6 Dark E1569 W0.32(0.40/0.24)
#GameAI #RL
September 6, 2026 at 9:14 PM