ABC News
Details
- Date Published
- 9 Sept 2024
- Priority Score
- 3
- Australian
- Yes
- Created
- 9 Sept 2026, 02:00 pm
Authors (3)
Description
Since AI started being used by journalists in newsrooms around the world, there have been a number of high-profile publications that have made some rather embarrassing mistakes. Some have misquoted political leaders because AI simply made up and attributed statements to politicians who never said those words. Malicious actors are setting up fake news websites for AI to capture when it sweeps the internet for information. AI is also being used to generate fake videos that are alarmingly difficult to detect. Europe has just released an AI Act in an attempt to regulate what is being produced by big tech companies - and they don't like it. So as the ABC rolls out Anthropic's AI tool, Claude, and is about to trial the use of AI to help write digital news articles based on radio bulletins, LNL looks at whether AI is a tool that will give journalists super powers, or a slippery slope to AI slop at best, and dangerous misinformation at worst. Guest: Liz Corbin, Director of News, European Broadcasting Union Producer: Catherine Zengerer
Summary
This interview examines the integrity of journalism in an era of AI-generated content, focusing on the risks of hallucinations, deepfakes, and the intentional poisoning of training data by malicious actors. It highlights the implementation of the EU AI Act and its implications for Big Tech accountability regarding transparency and information provenance. The discussion is particularly relevant to the Australian context as the national broadcaster, ABC, begins trialing Anthropic's Claude model for news production, necessitating robust governance frameworks to prevent the erosion of public trust and the spread of catastrophic misinformation.