RDLTR

Short for Read Later. Save links from any browser or phone. Click to save, read to clear. The full tour

One line. Many voicesSeek and you shall find

www.nytimes.com faviconWhy A.I. Safety Controls Are Not Very Effective

kept by

Jailbreaking AI systems to bypass safety controls remains trivially easy three years after ChatGPT launched, despite industry claims of progress.