ΑΙhub.org
 

GPT-3 in tweets


by
26 August 2020



share this:

AIhub | Tweets round-up
Since OpenAI released GPT-3, you have probably come across examples of impressive and/or problematic content that people have used the model to generate. Here we summarise the outputs of GPT-3 as seen through the eyes of the Twitter-sphere.

GPT-3 is able to generate impressive examples, such as these.

However, caution is needed when using the model. Although it can produce good results, it is important to be aware of the limitations of such a system.

GPT-3 has been shown to replicate offensive and harmful phrases and concepts, like the examples presented in the following tweets.

This harmful concept generation is not limited to English.

It is important to note that GPT-2 had similar problems. This EMNLP paper by Emily Sheng, Kai-Wei Chang, Premkumar Natarajan, and Nanyun Peng pointed out the issue.

GPT-3 should indeed be used with caution.

 




Nedjma Ousidhoum is a postdoc at the University of Cambridge.
Nedjma Ousidhoum is a postdoc at the University of Cambridge.




            AIhub is supported by:


Related posts :



An introduction to science communication at #AAAI2025

  18 Feb 2025
Find out more about our forthcoming training session at AAAI on 26 February 2025.

The Good Robot podcast: Critiquing tech through comedy with Laura Allcorn

  17 Feb 2025
Eleanor and Kerry chat to Laura Allcorn about how she pairs humour and entertainment with participatory public engagement to raise awareness of AI use cases

Interview with Kayla Boggess: Explainable AI for more accessible and understandable technologies

  14 Feb 2025
Hear from Doctoral Consortium participant Kayla about her work focussed on explanations for multi-agent reinforcement learning, and human-centric explanations.

The Machine Ethics podcast: Running faster with Enrico Panai

This episode, Ben chats to Enrico Panai about different aspects of AI ethics.

Diffusion model predicts 3D genomic structures

  12 Feb 2025
A new approach predicts how a specific DNA sequence will arrange itself in the cell nucleus.

Interview with Kunpeng Xu: Kernel representation learning for time series

  11 Feb 2025
We hear from AAAI/SIGAI doctoral consortium participant Kunpeng Xu.

The Children’s AI Summit – an event from The Turing Institute

  10 Feb 2025
Find out more about this event held ahead of the Paris AI Action Summit.
coffee corner

AIhub coffee corner: Bad practice in the publication world

  07 Feb 2025
The AIhub coffee corner captures the musings of AI experts over a short conversation.




AIhub is supported by:






©2024 - Association for the Understanding of Artificial Intelligence


 












©2021 - ROBOTS Association