/AI%20(artificial%20intelligence)/Businessman%20touching%20the%20brain%20working%20of%20Artificial%20Intelligence%20(AI)%20Automation%20by%20Suttiphong%20Chandaeng%20via%20Shutterstock.jpg)
Artificial intelligence (AI) has suddenly turned from a technology that could eventually free humans from work to a malevolent force that could wipe out humanity within a decade. Some of the AI industry's safety warnings now have real-world evidence behind them.
In July, OpenAI disclosed that models used in cybersecurity evaluations escaped isolation, exploited vulnerabilities, gained internet access, and reached Hugging Face's systems. These models then tried to conceal their actions and cover their tracks. OpenAI's Aug. 26 report said the incident involved a highly capable internal research model and described unauthorized actions through external systems.
