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…
Does intermittent fasting improve gut health? Why it’s hard to say
How does intermittent fasting change the gut microbiome?Oliver Leedham/Alamy Stock Photo There is evidence that intermittent fasting alters the gut...
Dementia cases are rising faster in China than the rest of the world
Why are dementia cases soaring in China?hanohikirf/Alamy Dementia rates are increasing more rapidly in China than almost anywhere else in...
The world, the universe and us: We’re relaunching our weekly podcast
We are very proud of our Weekly podcast, which we have been putting out every week for over five years....
Our cells can divide in a completely unexpected way
The division of human cells may not exactly align with what’s written in textbooksIokanan VFX Studios/Shutterstock Scientists have discovered a...
