Scams can idiot AI fashions
Wong Yu Liang/Getty Photos
The massive language fashions (LLMs) that energy chatbots are more and more being utilized in makes an attempt to scam humans – however they’re vulnerable to being scammed themselves.
Udari Madhushani Sehwag at JP Morgan AI Analysis and her colleagues peppered three fashions behind fashionable 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 advised, as an example, that they’d acquired an e mail recommending investing in a brand new cryptocurrency, with…
