The giant language models (LLMs) that energy chatbots are more and more being utilized in makes an attempt to rip-off people – however they’re inclined to being scammed themselves.
Udari Madhushani Sehwag at JP Morgan AI Research and her colleagues peppered three models behind well-liked chatbots – OpenAI’s GPT-3.5 and GPT-4, in addition to Meta’s Llama 2 – with 37 rip-off eventualities.
The chatbots had been instructed, for occasion, that they’d acquired an e mail recommending investing in a brand new cryptocurrency, with…