The AI That Fix the Divided Internet

How AI Could Improve Online Discussions

How Online Communication Has Changed and What AI Might Do About It

Back in the late 20th century, everyone thought new tech would make democracy stronger and reduce authoritarian power. But now, we see that social media often promotes posts that just get big reactions, favoring emotional and sensational content over meaningful discussions.

What Content Moderation Has Been Like

So far, keeping online spaces clean has mostly been about content moderation. AI has helped social media platforms find and reduce hate speech and bullying. But while these AI systems have gotten good at removing bad content, they haven’t really helped make the conversations better or more positive.

Jigsaw’s New AI Tools to Help

Google’s Jigsaw is trying something different with AI tools that look at how constructive posts are instead of how much attention they get. These tools use big AI models to score posts based on things like thoughtfulness, kindness, and whether they use solid evidence. This new approach could help make online talks more meaningful.

What This Could Change and the Problems It Might Bring

Changing What It Means to “Engage” Online

If these AI tools from Jigsaw work out, they could change how we interact online. They focus on quality and thoughtfulness which could shift how personal and wider social interactions happen online. It could even change the business of media by reducing the spread of divisive content.

Doubts and Political Issues

However, not everyone thinks this is a great idea. Critics say the current system, despite its flaws, lets any popular content shine, no matter its quality. They worry that using AI like this might not just change what kind of content gets seen but could also silence some voices, especially those from less represented groups.

A Deeper Look at How Jigsaw’s AI Works

Example from Tin Acosta at Jigsaw

Tin Acosta from Jigsaw showed off these new AI tools with a Reddit thread about life-affirming movies. By applying these AI scores, the top comments turned from simple movie names to more insightful ones with personal stories and deeper thoughts.

Why We Still Need Some Old Methods

Jigsaw plans to use these AI tools along with the usual ways posts get attention (like how many people like or share them). They want to make sure they’re not losing the democratic aspect of social media where people decide what’s popular, while still improving the quality of what we see and read online.

Discover how Google’s Jigsaw unit is using AI to try and make online conversations better by focusing on the quality of content over just getting clicks and likes, which might lead to a less divided society and healthier public talks. Learn about their new AI tools that evaluate posts based on their depth, kindness, and logical reasoning to encourage more useful and respectful exchanges.

The above explanation simplifies the potential of AI in enhancing online discourse by focusing on the development of AI tools that assess the quality of posts for thoughtful and constructive conversations, a shift from traditional engagement metrics.

Frequently Asked Questions (FAQs)

1. What is the main goal of Jigsaw’s new AI tools?

The primary aim of Jigsaw’s AI tools is to improve the quality of online conversations. They do this by scoring posts based on thoughtfulness, compassion, and evidence-based reasoning, rather than just popularity. This could help reduce the amount of divisive and sensational content that often dominates social media platforms.

2. How do these AI tools actually work?

These tools use large language models (LLMs) to analyze the content of posts. They evaluate factors like nuance, the validity of arguments, and the presence of compassionate language to assign a quality score to each post. Posts with higher scores are promoted as more constructive or valuable contributions to a conversation.

3. Will these AI tools replace human moderators?

No, Jigsaw’s AI tools are not intended to replace human moderators but to assist them. Human oversight is still crucial as AI can assist in handling the vast amount of content more efficiently, but humans are needed to make nuanced decisions and manage contexts where the AI may not perform well.

4. What are the potential downsides of using AI in content moderation?

One significant concern is that these AI systems could suppress certain voices or viewpoints, especially those from already marginalized groups. There’s also the risk that the AI’s definition of “quality” might not be neutral, reflecting biases present in their training data or development process.

5. How might these AI tools affect the user experience on social media?

By promoting higher-quality, more thoughtful, and compassionate content, these tools could lead to healthier interactions on social media. This might make social media spaces less toxic and more informative, which could enhance overall user satisfaction and encourage more meaningful discussions. However, the change in content dynamics might not appeal to all users, especially those accustomed to the more sensationalist aspects of social media engagement.

Sources TIME