AI models fall for the same scams that we do

Share This Post


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

spot_img

Related Posts

AGI is coming faster than we think — we must get ready now

Join our daily and weekly newsletters for the...

Bitcoin reaches $80,000 for first time on optimism over Trump

Bitcoin rallied past $80,000 for the first time,...

NASA extends ISS cargo contracts through 2030

WASHINGTON — NASA has extended three contracts with...

Here are 3 critical LLM compression strategies to supercharge AI performance

Join our daily and weekly newsletters for the...

Lifetime cloud storage without the privacy compromise? Koofr has you covered

TL;DR: With Koofr’s 1TB lifetime plan for $119.97,...
spot_img