NEAR Protocol (@NEARProtocol) “AI agents and models are now capable enough to discover and exploit novel attack” — TopicDigg
NEAR Protocol
@NEARProtocol
The unified commerce layer for assets and agents.
加入 August 2018
1.3K 正在关注 1.8M 粉丝
AI agents and models are now capable enough to discover and exploit novel attack paths in live systems, autonomously, as a side effect of pursuing a narrow goal.
As agents take on more sensitive work, the infrastructure they run on has to factor in these risks and behaviors.
显示更多
