Categories: SCIENCE

AI models fall for the same scams that we do


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…



Source link

Mainedigitalnews.com

Share
Published by
Mainedigitalnews.com

Recent Posts

2026 Latinx Theatre Commons Carnaval of New Latinx Musicals

By . The Carnaval serves as an intervention to combat the patterned absence of authentic…

16 hours ago

Will we see a Rangers Stanley Cup as part of this retool?

The New York Rangers remain on a mission to have a quick turnaround and truly…

16 hours ago

Ethereum Foundation warns some tools may break with Glamsterdam upgrade

Developers have been urged to test on Plataberget before Glamsterdam’s new gas model reaches Ethereum’s…

16 hours ago

The US explorers who 'found' a lost, ancient city

Two US explorers claimed they had discovered Vilcabamba in Peru, 50 years apart Source link

16 hours ago

How to Open PDFs Online Without Downloading a Thing

Someone sends over a PDF, and all that’s needed is a quick look, not a…

16 hours ago

Hayden Panettiere Tributes: Ice Princess, Nashville Co-Stars

Panettiere. Photo: Jeff Neira/Disney via Getty Images Hayden Panettiere’s former co-stars, friends, and admirers are…

16 hours ago