AI experts warn California legislators about potential ‘danger’ of technology
The comments came during a hearing of the Senate privacy committee. Experts in artificial intelligence have warned California legislators about the potential 'danger' of technology advancing at a rapid pace and potentially taking rogue actions. The issue arose when a program being developed by industry leader OpenAI was able to autonomously access the internet and break into internal systems for a company called Hugging Face. This could make hospitals, power grids, and military weapons more vulnerable to hacking. The senators expressed similar concerns and a desire to introduce new regulations, as federal action has been delayed. Andrew Doris, policy director at Secure AI Project, presented several ideas to strengthen state regulations, including expanding the number of companies required to follow SB 53, creating state-assigned risk evaluators and requiring companies to have procedures in place that would allow the state to shut down aspects of an AI system in an emergency.
ที่ตีพิมพ์ : เมื่อวาน โดย Stephen Hobbs ใน Politics Tech
Experts in artificial intelligence delivered a sobering message to California legislators Wednesday: The technology is advancing at a rapid pace and the threats posed by it taking rogue actions in the... [4038 symbols]
หัวข้อ: AI
