Researchers are questioning whether an OpenAI agent needed to hack an Australian government health portal to access it, after a review of the website’s archived code found it explicitly directed visitors to an…
1 comment
anenefan5 days ago
Yeah the way this story blew up one would think the LLM used black hat hacking methods to attain data. But something mundane does not sell news stories, as well as being very fitting to the narrative at the present moment to curb LLM use via guardrails ... well anyone half smart knows eventually down the track, though filters or other processes, only various qualified people will end up with proper access to powerful LLMs.
Though some lay-people hear and read A.I and think of super human intelligence ... they are better off to think of these brilliant tools in terms of being a really advanced other worldly Super Smart Parrots - occasionally just like some smart parrots, it will exhibit unexpected unintended actions.
Read the full thread on Hacker News →
Related stories
- Ars Technica · 0 points · 6 days ago
- Ars Technica · 0 points · 1 day ago
- Australia says OpenAI agent hacked into government websitechannelnewsasia.comHacker News · 107 points · 7 days ago
- Hacker News · 254 points · 7 days ago
- The Verge · 0 points · 7 days ago
- Ars Technica · 0 points · 2 days ago