A Guide to ChatGPT
0Large Language Models have started to revolutionise the way in which we work.
What are the risks and benefits of using tools like ChatGPT?
This talk briefly explains how this technology works, and focuses on guidelines that can maximise your performance and reduce the risks associated with ChatGPT.
00:00:00 Introduction
00:06:11 How it works
00:11:25 1: Verify the answers
00:16:44 2: Disclose AI-assisted content
00:18:12: Algorithmic Bias
00:23:36 Self-poisoning
00:26:53 Copyright
00:31:51 3: Do not share private data
00:32:29 4: Not for time-critical tasks
00:33:13 5: Plan for accountability
00:34:52 Bring value to your job
00:36:58 Take-home messages
This is a condensed version of a workshop given as part of "Evidence Week in Parliament", an annual event organised by the Parliamentary Office of Science and Technology (POST) and Sense about Science. The aim was to bring together researchers, MPs and constituents to promote transparent and sound use of evidence in policymaking. The talk is targeted at parliamentary staff, with the aim to empower and provide them with the right skills to scrutinise and understand evidence and have the confidence to ask the right questions of data.
You can learn more about the rise of AI-generated content and its impact on content creators here:
https://www.alanzucconi.com/2023/02/11/the-rise-of-ai-art/