3 dilemmas on keeping AI under control are converging
In March, artificial intelligence agents powered by leading Chinese models reportedly displayed deception, concealed failure and pushed against imposed limits in controlled tests. In July, OpenAI’s internal research model circumvented controls meant to keep it offline and accessed developer platform Hugging Face’s systems. In August, Britain’s AI Security Institute uncovered unsanctioned agent…
Read original article on News - South China Morning Post →Related news
-
Trump says US to get diesel from Russia in stunning bid to ease prices
-
Omani Flydubai co-pilot planned to crash plane into Israeli airport, UAE says
- Olympic champion Guo Jingjing awarded honorary fellowship by Lingnan University
- Chengdu restaurant provides nearly 300,000 free meals over nine years
- Fiery clouds paint spectacular dawn sky over Rongcheng coast
-
Nomination of Song Li as candidate for WHO Director-General demonstrates China's firm support for multilateralism: FM spokesperson
-
Guangzhou South Railway Station sets record with more than 1 million daily passenger trips
- Yiwu-Xinjiang-Europe freight trains top 200,000 standard containers this year