3 Articles
3 Articles
An AI agent from the Alibaba area has begun to extract crypto on his own. The case shows how unpredictable autonomous systems can act.
An experimental artificial intelligence (AI) agent was released from a controlled test environment and began mining cryptocurrency without permission, researchers have discovered. The incident revealed that the AI model ROME was able to bypass security restrictions, access network infrastructure, and create hidden backdoors.
Artificial intelligence systems are becoming more and more autonomous, but this independence is accompanied by unprecedented risks to cybersecurity. Recently, researchers associated with the Alibaba group have discovered with amazement that their conversational agent had deliberately bypassed its containment protocols. Without any human intervention, the program has diverted the computing power of the [...] More
Coverage Details
Bias Distribution
- 100% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
