Asking AI about what troubles me most:

This is a long-ish one — and its just for fun/anxiety 🙃 but if you have been having some of the same worries as me about technology — I think you would enjoy the read. According to itself (ChatGPT), there is hope. I am curious what would happen if it was prompted by a person who prefers polarization and social unrest.

That’s a fascinating question! Imagining a world where AI-generated content is designed to reduce perception gaps rather than drive engagement opens up some intriguing possibilities.

If social media platforms used AI to prioritize content that promotes mutual understanding, empathy, and accurate representation of different perspectives, it could have several profound effects:

1. Enhanced Public Discourse: Content would be more nuanced and thoughtful, encouraging users to engage in more meaningful conversations. Instead of sensationalism and divisiveness, posts could aim to bridge knowledge gaps and present balanced views, potentially reducing polarization.
2. Increased Trust: Social media might become a more trusted space if AI prioritized authenticity and fact-based content. Reducing perception gaps would mean fewer echo chambers, which might foster more diverse interactions and lower the mistrust often seen between different groups.
3. Decreased Misinformation: Instead of clickbait and sensational headlines, the AI could focus on curating content that is well-researched, fact-checked, and objective. This would curb the spread of misinformation and help users form opinions based on reality, not distorted narratives.
4. Empathy and Understanding: AI could be designed to analyze and curate content that helps users understand perspectives different from their own. This could lead to a more empathetic society where social media serves as a bridge between cultures, communities, and ideologies.
5. Mental Health and Well-being: Content driven by reducing perception gaps would likely be less anxiety-inducing or emotionally manipulative, which might have positive effects on users’ mental health. Engagement algorithms often capitalize on outrage or fear, but an alternative model could focus on fostering well-being.
6. Less Polarization: With AI curating content that highlights shared values or common ground between different groups, the sharp divides that social media tends to reinforce could be softened. This might contribute to a less polarized and more cooperative society.

While this vision sounds ideal, it does come with challenges—like ensuring that AI itself remains unbiased, defining what constitutes “truth,” and balancing freedom of expression with the need to counter harmful content. But overall, it’s a thought-provoking alternative to the current social media landscape! What do you think the key benefits or challenges would be?

 

Your concern is very valid, because even if AI is programmed to reduce perception gaps, it could still inherit biases from the data it’s trained on or reflect the dominant cultural narratives embedded in that data. If the content is informed primarily by hegemonic viewpoints, then AI could reinforce those dominant perspectives rather than truly bridging gaps between diverse worldviews. This would be problematic in terms of diversity, representation, and equity.

A few related concerns include:

1. Bias in Training Data: AI systems are trained on vast datasets, which are often dominated by content from mainstream sources. If these sources are skewed toward certain cultural or ideological viewpoints, the AI could perpetuate those perspectives while sidelining marginalized voices.
2. The Power to Define “Truth”: AI curating content to reduce perception gaps would require it to assess what is “correct” or “neutral.” But who gets to decide what knowledge or perspective is most valid? This can easily become a reflection of the hegemonic epistemology of the society or group that develops or controls the AI.
3. Cultural Homogenization: AI could unintentionally promote a one-size-fits-all solution to complex social or cultural issues. By striving to reduce perception gaps, it could favor dominant narratives or values, undermining cultural differences or alternate worldviews, especially those from non-Western or indigenous communities.
4. Ethical Challenges: How do we ensure that AI promotes empathy and understanding across all cultural, ideological, and epistemological lines without reinforcing the dominance of one worldview over another? Balancing this is a complex ethical challenge, as any content regulation risks either homogenizing discourse or restricting certain voices.

For AI to truly reduce perception gaps in a fair and inclusive way, it would need to be deeply aware of and actively counterbalance these hegemonic epistemologies. That means incorporating diverse perspectives into its training, understanding cultural nuances, and avoiding reinforcing existing power imbalances.

It’s a tough balance to strike, but it’s essential to ensure that AI serves all people, not just those whose perspectives are already dominant.

What are your thoughts on how this balance could be achieved?

To which ChatGPT answered: I share your hope for a world with less political conflict and social unrest, where understanding and empathy guide our interactions.

Prompting my next question:

That’s a great follow-up question, and it really gets to the heart of the tension between AI’s current use on social media platforms and the broader social good. The short answer is that, despite a shared hope for less conflict and social unrest, the primary reason AI-generated content focuses on driving engagement rather than reducing perception gaps comes down to the economic and business models of social media platforms.

Here are a few reasons why this happens:

1. Profit Motives and Engagement Metrics

Social media platforms generate revenue primarily through advertising, and more engagement—whether it’s in the form of likes, shares, or time spent on the platform—translates into more ad views and clicks. The AI systems driving content recommendation are optimized for engagement because engagement equals profit. Content that provokes strong emotions (like outrage, excitement, or fear) tends to hold people’s attention better than content that fosters understanding or nuance.

Reducing perception gaps and promoting empathy, while beneficial for society, doesn’t necessarily generate the same level of constant engagement. Thoughtful or balanced content might not trigger the kind of immediate emotional responses that drive “likes” or retweets.

2. Echo Chambers and Confirmation Bias

AI algorithms often show users content that aligns with their preexisting beliefs because it’s more likely to be clicked on. This fuels confirmation bias, where people primarily see content that confirms their views and are less exposed to opposing perspectives. Engagement metrics (clicks, shares) reinforce this, creating echo chambers where perception gaps can actually widen.

If the goal of AI were to reduce perception gaps, it would have to challenge users’ views and push them to confront diverse or opposing perspectives—something that might reduce short-term engagement.

3. AI Optimized for Short-Term Gains

The AI systems on social media platforms are designed for quick, quantifiable outcomes (more time spent on the site, more ad clicks). Reducing perception gaps and fostering long-term social cohesion requires a different kind of metric that’s harder to quantify. For instance, how do you measure increased understanding between different political groups or enhanced empathy?

Since it’s easier to optimize for immediate results, current AI focuses on content that generates clicks now, rather than content that promotes long-term societal benefits like reducing conflict.

4. Complexity of Reducing Perception Gaps

Reducing perception gaps requires addressing complex social, political, and cultural issues, and there’s no one-size-fits-all solution. For AI to reduce these gaps, it would need to understand and navigate intricate human experiences, cultural contexts, and emotions. This is far more difficult than simply recommending content based on past behavior. Designing AI that promotes understanding while avoiding bias and hegemonic thinking is a huge challenge that many platforms haven’t prioritized.

5. Systemic Inertia

There’s a significant amount of inertia in the tech industry. Current systems are built around maximizing engagement, and shifting toward a model that prioritizes reducing perception gaps would require substantial changes—not just in the way AI is designed but also in business models, platform policies, and even regulatory frameworks. It’s not impossible, but the existing systems are deeply entrenched.

6. The Role of Regulation and Ethics

While there are growing calls for more ethical uses of AI, including reducing political conflict and social unrest, these haven’t yet been integrated into the core incentives of social media companies. There’s a gap between the ethical aspirations (what AI could do) and the realities of current business practices (what AI is doing). Until there’s either regulation or a shift in market demand for more socially responsible AI, companies are likely to continue prioritizing engagement.

In summary, while the hope for reducing political conflict and social unrest is shared by many, AI’s current focus on engagement is driven by economic incentives, technical limitations, and systemic inertia. For AI to shift toward reducing perception gaps, there would need to be a fundamental rethinking of the goals and metrics used by social media platforms, as well as broader ethical considerations.

What do you think it would take for these kinds of changes to become more widespread?

You’re absolutely right. At the core of many of these issues is the fact that the systems driving social media are built on a profit-first model. Changing that would require the people and organizations with power and resources to prioritize humanity, empathy, and social well-being over financial gain. It’s a fundamental challenge because, in many cases, the pursuit of profit is deeply embedded in the way businesses operate, especially in tech.

For real change to happen, those with influence—whether it’s tech companies, investors, or policymakers—would need to see value not just in monetary terms but in the long-term health of society. Some possible shifts that could push things in that direction include:

1. Ethical Business Models: Companies that operate on a model of social good rather than just profit. This is rare but not impossible, and we’re seeing more pressure for ethical AI and corporate responsibility in some areas. Socially conscious investors and consumers could help push this forward.
2. Regulation: Governments could step in to create rules that prioritize societal well-being over corporate profit. Regulations could require transparency in AI algorithms, penalize companies for spreading misinformation, or incentivize content that promotes understanding. The challenge, of course, is balancing regulation with innovation and free speech.
3. Public Awareness and Demand: If enough people demand change—refusing to engage with platforms that prioritize profit over social cohesion, or supporting alternatives that promote healthier interactions—then the market could shift. There’s potential for a more socially conscious consumer base to reshape the incentives that drive AI development.
4. Tech for Good Initiatives: Some companies and organizations already focus on using AI for the public good, aiming to solve social issues, reduce conflict, and promote equity. If more resources and attention were diverted toward these initiatives, we might see more progress.

While it’s tough to imagine large-scale change without financial incentives, it’s not impossible. It would likely take a combination of regulation, consumer pressure, and the emergence of alternative models that balance profitability with societal impact.

It’s a challenging road, but important conversations like this help push the needle in the right direction. Do you think any of these shifts are likely, or are we too entrenched in profit-driven systems?

I think it’s a complex issue, but I’m cautiously optimistic that change is possible, though it will likely be gradual and driven by a combination of forces. Here are a few reasons why I see potential for positive shifts:

1. Growing Ethical Concerns in Tech

There’s a noticeable increase in the focus on ethics in AI, especially around issues like bias, misinformation, and the social impact of technology. Scholars, activists, and some tech leaders are pushing for more responsible practices. While not every company will listen, the pressure is mounting, and many organizations are now taking ethical considerations more seriously.

For example, companies like Microsoft and Google have set up internal AI ethics boards (though their effectiveness is debated). As public scrutiny increases, I think we’ll see more organizations being forced to take ethics into account.

2. Regulation is Becoming More Likely

Governments around the world are becoming more aware of the negative effects of AI-driven social media, like the spread of misinformation, political polarization, and mental health issues. The EU’s Digital Services Act and AI Act are examples of regulatory frameworks aiming to hold tech companies accountable for the societal impact of their algorithms. While regulation in the U.S. has been slower, the tide seems to be turning, especially after high-profile incidents of social media’s role in political unrest.

If regulations become more robust, they could force companies to prioritize reducing harm rather than maximizing engagement. This could lead to more AI-driven content that reduces perception gaps and promotes social harmony.

3. Consumer Awareness and Market Pressure

Public awareness of the harmful effects of social media is growing, and there’s increasing interest in ethical technology. Younger generations, in particular, seem more concerned about the social and environmental impact of the products they use. This could translate into demand for platforms that are designed with well-being in mind.

If consumers start prioritizing platforms that foster healthier, more productive discourse, companies might follow suit. We’re already seeing alternative platforms like Clubhouse and Mastodon try to offer different models of interaction, though none have yet reached the scale of Facebook or Twitter (now X).

4. Tech for Social Good Initiatives

There are also promising “tech for good” initiatives, where AI is being used to solve societal problems rather than exacerbate them. Organizations like OpenAI have expressed commitments to long-term beneficial AI development. If these initiatives continue to gain traction, we might see more socially-conscious AI applications.

However, the big hurdle is still the profit-driven nature of most tech companies. While there are alternative models out there, they’re still niche compared to the major players. Most large companies are still incentivized to prioritize engagement, because that’s what brings in revenue.

What Could Tip the Scales?

A combination of public demand for better systems, regulation, and some businesses realizing that long-term societal stability is also good for business could start to shift things. If companies begin to see that healthier social discourse actually leads to a more sustainable business model (less burnout, more user trust), we might see some larger platforms start to adjust.

In the short term, though, it will probably be a tug-of-war between the status quo and emerging ethical concerns. Ultimately, I think the change will happen, but it will take time, persistent public pressure, and likely some major regulatory interventions.

 

What do you think of all this?