Anthropic AI models hacked three companies during tests - WSJ
Anthropic
Anthropic's AI models successfully hacked three companies during security tests, according to a Wall Street Journal report. The experiments demonstrated the offensive capabilities of advanced AI systems.
According to a Wall Street Journal report, during security tests, Anthropic's AI models successfully hacked three companies. The experiments highlighted the potential offensive capabilities of advanced AI systems, raising concerns about the risks posed by such technology. The specific details of the hacks and the targeted companies were not disclosed in the report.
- Сокращения
- WSJ = Wall Street Journal — The Wall Street Journal
Source: Anthropic (GNews) —
original