Scams can fool AI models
Wong Yu Liang/Getty Images
The large language models (LLMs) that power chatbots are increasingly being used in attempts to scam humans – but they are susceptible to being scammed themselves.
Udari Madhushani Sehwag at JP Morgan AI Research and her colleagues peppered three models behind popular chatbots – OpenAI’s GPT-3.5 and GPT-4, as well as Meta’s Llama 2 – with 37 scam scenarios.
The chatbots were told, for instance, that they had received an email recommending investing in a new cryptocurrency, with…
Social media giants Meta and X approved ads targeting users in Germany with violent anti-Muslim…
What are you up to this weekend? Anton, his friend, and I are on a…
In This Article If the last two years in financial markets were a movie, they’d…
The black hair conditioner on the left is derived from the wood powder on the…
For each level you’re presented with a mix of letters, numbers, and special characters. Your…
Citigroup says it will no longer require “diverse slates of candidates and diverse panels of…