Kradle: Eval AI with Simulations

Posted4 months agoActive4 months ago

ivolo

11 points

5 comments

twitter.comTechstory

excitedpositive

Debate

20/100

AI SafetySimulationsMachine Learning

Key topics

AI Safety

Simulations

Machine Learning

Kradle is introducing a platform that uses simulations to evaluate AI models, sparking discussion on its potential applications and overlap with existing approaches.

Snapshot generated from the HN discussion

Discussion Activity

Light discussion

First comment

36m

Peak period

0-1h

Avg / period

Key moments

01Story posted
Sep 8, 2025 at 3:02 PM EDT
4 months ago
Step 01
02First comment
Sep 8, 2025 at 3:38 PM EDT
36m after posting
Step 02
03Peak activity
5 comments in 0-1h
Hottest window of the conversation
Step 03
04Latest activity
Sep 8, 2025 at 3:56 PM EDT
4 months ago
Step 04

Generating AI Summary...

Analyzing up to 500 comments to identify key contributors and discussion patterns

Discussion (5 comments)

Showing 5 comments

alexisgauba

4 months ago

1 reply

very cool. how are ya'll thinking about genie 3?

jamest

4 months ago

Genie 3 is super impressive -- but you can't manipulate it programmatically like Minecraft.

We've built our infra so that we can plug in any simulation environment. If an AI-generated world starts being programmatically modifiable (and has really solid object permanance :) then we'd happily use it!

jamest

4 months ago

Hi HN -

Minecraft has a DSL that lets you manipulate the world. We've piggy-backed on that, along with a K8s infra to run N worlds in parallel, to let you create simulations of arbitrary complexity.

We think simulations are the best way to test frontier AIs due to their degrees of freedom and expressivity.

AMA!

stopachka

4 months ago

Interesting work! This can be a great way to have ML models live in a sandbox and prove out some of the safety concerns Anthropic et al speak about.

benhylak

4 months ago

how do you see the world-simulation approach overlapping with the sort of custom-RL environments (e.g. mock Salesforce apps) that people are building out for frontier labs?

similarly, do you see it as a general test of intelligence? more for robotics?

View full discussion on Hacker News

ID: 45172410Type: storyLast synced: 11/20/2025, 8:37:21 PM

Want the full context?

Jump to the original sources

Read the primary article or dive into the live Hacker News thread when you're ready.

Open link View on HN