AI chatbots become more sycophantic as they get more advanced – New Scientist

Advertisement
Explore by section
Explore by subject
Explore our products and services
If a person says they believe an objectively false statement, AIs tend to agree with them – and the problem seems to get worse as models get bigger
By Matthew Sparkes
18 August 2023

AI models have a tendency to tell us what we want to hear
Reismedia/Shutterstock
Artificial intelligence chatbots tend to agree with the opinions of the person using them, even to the point that they nod along to objectively false statements. Research shows that this problem gets worse as language models increase in size, adding weight to concerns that AI outputs cannot be trusted.
Jerry Wei at Google DeepMind and his colleagues ran experiments on AI models with 8 billion, 62 billion and 540 billion parameters – the values that each model …
Advertisement
To continue reading, subscribe today with our introductory offers
No commitment, cancel anytime*
Offer ends 14th June 2023.
*Cancel anytime within 14 days of payment to receive a refund on unserved issues.
Inclusive of applicable taxes (VAT)
Existing subscribers
Advertisement
Explore the latest news, articles and features
News
Subscriber-only
Culture
Subscriber-only
News
Subscriber-only
News
Subscriber-only
Trending New Scientist articles
1
2
3
4
5
6
7
8
9
10
Advertisement
Download the app

source