Bad science: AI used to target kids with disinformation on YouTube - BBC World Service

Summary

The video script discusses the rise of AI-generated 'bad science' content on YouTube, misleading children with false information. It highlights how these videos exploit algorithmic recommendations and outperform genuine educational content, potentially harming children's understanding of science. The script also includes an experiment showing the rapid spread of misinformation and the need for greater scrutiny and regulation.

Takeaways

• 📺 Under 18s are the primary audience on YouTube, making them susceptible to the content they encounter.
• 🎮 Popular YouTubers like DanTDM and Mr. Beast have significant influence, but new AI-generated content is emerging.
• 🐾 There is a growing concern about AI-generated videos spreading false information, particularly in the realm of science education.
• 🌐 These misleading videos are optimized for YouTube's algorithm, often featuring catchy titles and controversial topics.
• 🧬 Research indicates that creators exploit children's clicks for profit while spreading misinformation in classrooms worldwide.
• 🔬 Misleading videos are tagged as 'educational content,' often outperforming legitimate science videos in recommendations.
• 🌐 An experiment showed that 'bad science' channels quickly appeared in recommended videos after watching legitimate science content.
• 🌍 The issue is global, with channels translating misleading videos into over 20 languages, potentially affecting a wide audience.
• 👦👧 Children from different countries, such as the UK and Thailand, were shown to be influenced by these misleading videos, believing in false claims.
• 🤖 AI is used to generate scripts, voices, and footage for these videos, making it easier for anyone to create misleading content quickly.
• 💰 The financial incentive is significant, with misleading channels earning thousands through advertising revenue, and YouTube taking a cut.

Q & A

What is the primary audience on YouTube?

Under 18s are the biggest audience on YouTube.

What type of content is being created using Artificial Intelligence that is causing concern?

A new type of content creator is using Artificial Intelligence to create videos full of false information, often tagged as 'educational content'.

How do these AI-generated videos exploit children's clicks?

These creators take advantage of kids’ clicks to cash in, spreading misinformation to classrooms around the world.

What is an example of a false claim made in these 'bad science' videos?

One example is that pyramids can make electricity, which is a false claim presented as fact in these videos.

How are these videos optimized for the YouTube algorithm?

These videos are optimized with catchy titles and controversial topics to attract more views and engagement.

What was the outcome of the experiment conducted to see if these videos were reaching children?

After only four days, one of the 'bad science' channels appeared in the recommended videos for the children's accounts set up for the experiment.

How do these AI-generated videos affect the credibility of legitimate science content?

These videos often beat legitimate science videos in the race to be recommended, potentially pushing good information out and drowning out good content.

What is the process of creating an AI-generated video?

A video needs a script, which can be generated in seconds with AI. Then it needs a voice, which can be synthesized, and AI can find footage from across the internet to piece together the final film.

How do these channels manipulate footage from legitimate educational creators?

They take footage and graphics from legitimate educational creators, change or ignore the original meaning, and repurpose them into false information.

What is YouTube's response to the issue of misinformation on their platform?

YouTube recommends YouTube Kids for under 13s, which has a 'higher bar' for videos shown, and they are committed to removing misinformation from their platforms. They also provide information panels that show 'additional context' on conspiracy-related content.

How can parents and educators counter the spread of misinformation from these AI-generated videos?

As teachers and parents, they need to have conversations with children about what they're watching and the media they're absorbing, and encourage them to think critically and question the content.