Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
claytonia
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
claytonia
5mo ago
This looks less like a model issue and more like a systems design problem. A vague objective (“make money”) without constraints, budgeting, or basic rules will naturally lead to this kind of behavior. The agent just optimizes whatever it wa
2.
▲
by
claytonia
6mo ago
Nice project. "Pattern-based injection detection is easily bypassed by obfuscated inputs" is a sharp finding. It got me thinking about two different ways to do agent security. What you're building is content security: scan th
3.
▲
by
claytonia
6mo ago
Cool demo, but letting an AI Agent drive a physical probe may have problems. AI is probabilistic, but hardware is precise. If the model miscalculates a pin's position by even 0.1mm, the probe may crush the board. I am curious how the a
4.
▲
Introducing Unity Machine Learning Agents
(blogs.unity3d.com)
13 points
by
claytonia
9y ago
|
2 comments