The AI Construction of Reality
How ChatGPT can help us build a shared reality and bring us together
There are many explanations for why we are so polarized in American society. Some argue that people have burrowed into ideological media bubbles and only consume ideas that make their “side” look good and the other “side” look bad. Others have argued (as I have done here) that we no longer have an array of social identities to choose from. We can longer say, “I am a Republican but also a Christian and also a veteran and also a second-generation Italian immigrant,” and then find commonalities with others along one of these three dimensions. No. Now many people have only one identity to choose from - liberal or conservative- creating a binary us versus them dynamic. Still, others argue that we have too much geographic segregation - liberals live in their cities, and conservatives live in theirs.
What ties all of these explanations together, in my mind, is the notion that the worlds of liberals and conservatives are constructed so differently to the point where they no longer share the same reality.
The Social Construction of Reality
When sociologists claim that our reality is socially constructed, they are probably referring to the claim that our understanding and experiences of the world, including our perceptions, knowledge, and beliefs, are not solely determined by objective facts. They are shaped by the people and institutions we interact with within a specific social and cultural context.
That is a mouthful. But the idea is simple and self-evident once it is explained.
All the information we take in is through our senses. We gather as much information about the world as possible by touching, hearing, tasting, smelling, and seeing. Imagine you are sitting in a family room or a kitchen. You see a coffee mug on the kitchen counter. You smell barbecue wafting in from the backyard.
But we don’t simply catalog this information in a value-neutral, robotic way.
That coffee mug is not simply a cylindrical container. It is the coffee mug your mother uses every morning to drink coffee. That cylindrical container has been in your mother’s hands on many a day as you two have argued, laughed, and cried together. Your experiences with your mother have shaped how you understand and experience that coffee mug.
The smell of barbecue? Well, you are not only recognizing the smell of meat cooked by charcoal briquets. Instead, the idea of a barbecue is likely associated with a cookout. Think of the Fourth of July (America's most popular cookout day). Barbecuing is an ancient practice done globally, but if you are an American, the smell of barbecue comes with expectations. We have an expectation that we should cease economic activities on this particular day, convene with family and friends, and cook meat like ground beef formed into a disc or ground pork encased in sheep’s intestine. We may even expect to watch small packets of gunpowder explode around us at night. These expectations are in part constructed by the institutions in our society - the government’s passing of a law making the 4th of July a national holiday and businesses advertising cookout products in magazines. The smell of grilled meat cannot be completely disentangled from the constructed association of barbecuing, cookouts, vacations, fireworks, and the Fourth of July.
The coffee mug and the barbeque are examples of how the people and institutions we interact with construct our reality. In most cases, this is a benign aspect of human existence. So what if one’s partner does not see the significance of that old chipped coffee cup you are keeping or that your Bangladeshi neighbors look on curiously as you haul out your grill on the 4th of July and overcook some ribs?
But it matters greatly when it comes to how we understand our government and institutions.
I have had conversations with people online who seem to live in an entirely different world. Their realities have been constructed differently than mine. The differences in construction could be because of differences in media consumption, social media connections, or the neighborhood one lives in. Maybe all three. But at the end of those differences are realities that are vastly different.
Many people earnestly and confidently believe that the Covid Vaccine is unsafe and that a liberal media and pharmaceutical industry was lying to them or forcing them to take something they didn’t need to control them.
My world was not constructed in that way. I am one of those (regrettably) stereotypical men who loathe going to the doctor, and I had no interest in taking the vaccine. But I did not, at any point, think there was some plot to infect me with a damaging substance or that the government was inching towards taking over my life. I saw the vaccine as a testament to how humanity can come together and work to solve a serious crisis!
The differences here are not about isolated facts - but facts that are woven together into meaningful narratives. It is not simply: “Is the vaccine safe” or “Can the vaccine spread through the air?” It is a narrative that weaves together claims about the origin and safety of the vaccine with why it is being deployed by the federal government to produce a coherent story.
And so it goes with so many other issues.
Well, my reality is that this esoteric body of thought is not in schools, but even if it was…so what? Maybe it helps older students critique the laws and policies that structure their lives. Other people’s reality - CRT is being taught everywhere, and it teaches young people that white people are sinful and racist.
Young people transitioning?
My reality is that some young people's gender identity does not align with their sex. I believe professionals who specialize in or work with trans persons have the best interest of these young people in mind and advise them the best they can. Other people’s reality - these young people are either being influenced by older trans persons on Instagram or suffering a mental illness, and health professionals, trapped in the throes of trans ideology, are pushing drugs and surgeries on them without a proper assessment.
Different worlds, man.
Maybe there is a way out of this.
I am working with a student this summer who wants to explore ChatGPT. Until they approached me, I had heard of the technology but never really tried it.
When I did, I was impressed with the answers it gave. For example, above, I talked about CRT. If I ask ChatGPT, “Is CRT in schools?” I get a succinct, even answer. I quote at length, and if you already know the answer, you may want to skip ahead:
While CRT itself is not typically included as a curriculum subject in K-12 education, discussions about incorporating concepts related to race, racism, and social justice into curriculum frameworks have become more prevalent in recent years. Some educational initiatives and policies aim to address issues of systemic racism and promote a more inclusive and comprehensive understanding of history and society.
However, the extent and manner in which concepts related to CRT or broader discussions of race and racism are included in school curricula can vary significantly across different jurisdictions, school districts, and educational contexts. Specific decisions about curriculum are typically made at the local level, taking into account local educational standards, community values, and legal considerations.
It is important to engage in nuanced and informed discussions about how race and racism are addressed in educational settings. Supporters argue that including discussions of race and racism can help foster a more inclusive and equitable learning environment. Critics may express concerns about potential ideological bias, age-appropriateness, and the role of education in shaping students' beliefs.
Given the contentious nature of the topic, it is recommended to engage with multiple perspectives, rely on reputable sources, and participate in constructive discussions to gain a comprehensive understanding of how race and racism are addressed in specific educational settings.
This is just excellent. It makes the claim that CRT is typically not in schools - which is correct based on all I have read and all the people I have talked to. It then adds nuance by making an additional claim that I think is correct as well - while CRT proper is not in schools, concepts related to race, racism, and social justice are being discussed like never before. I also like the third paragraph, which discusses the views of both critics and supporters of CRT-related instruction.
Seriously this is awesome.
What if everyone had their reality about CRT constructed in this way? What if when everyone hears the word “CRT,” the meaning includes what is narrated above? We wouldn’t have people like Chris Rufo and James Lindsay taking advantage of people’s ignorance and generating false narratives that sacrifice truth for political expediency.
So how does this technology work?
Well, I asked ChatGPT to describe itself:
ChatGPT is a deep learning model that uses a transformer architecture to generate human-like text based on the input it receives. The model is trained on a large amount of text data and learns patterns and relationships within the language, allowing it to generate coherent and contextually relevant responses.
The "Chat" in ChatGPT signifies that this particular implementation is designed for interactive conversational purposes. It is trained on a wide range of conversational data and can respond to user prompts and questions in a conversational manner.
Overall, ChatGPT is an AI language model designed to engage in text-based conversations and provide relevant and informative responses based on the input it receives.
Some of this is goobledy-gook to me. But I believe I get the gist. ChatGPT uses prior conversations produced online about a topic to answer present questions about that same topic. I suppose many people have asked about CRT and had many conversations online about the topic, and the answers from ChatGPT are a summation of what is commonly said. It is a kind of collective wisdom or Internet zeitgeist.
Can ChatGPT Avoid Being Slandered “Woke”?
There is a potential for ChatGPT and other AI technologies to provide a non-partisan and value-neutral source of information for people.
But looking forward, the hurdle will be those who refuse to accept the technology because the reality it constructs too often goes against their political interests.
For example, when I asked ChatGTP, “Was the 2020 presidential election stolen” I got this:
“No, there is no credible evidence to support the claim that the 2020 presidential election in the United States was stolen. The 2020 election underwent extensive scrutiny, including recounts, audits, and legal challenges, and multiple courts across various states rejected allegations of widespread voter fraud or irregularities that could have impacted the outcome of the election.
The Cybersecurity and Infrastructure Security Agency (CISA), a federal agency responsible for securing the nation's critical infrastructure, described the 2020 election as the "most secure in American history." Additionally, the Department of Justice found no evidence of widespread fraud that could have affected the outcome of the election.
Claims of a stolen or rigged election emerged primarily through misinformation and unfounded conspiracy theories. These claims have been thoroughly investigated and debunked by numerous independent fact-checking organizations and reputable news outlets.”
This kind of response is anathema to people determined to believe that the 2020 election was rigged. They will try and sow mistrust around the technology - maybe even calling it “woke.” Some, including Dr. Jordan Peterson are already working towards this goal. See Jordan Peterson below having problems with the answer to the question “what is a woman” from Google’s AI chatbot Bard.
But unlike people and groups’ ability to sow mistrust in our educational and government institutions, it will be more challenging with AI.
Young people? Forget it. They will use AI to do their homework assignments, help them write resumes, help them be content creators, and more. They will trust it in the same way they were the first to trust the Internet and Wikipedia.
I also think most other people won’t be so easily jaundiced towards the new technology. It is hard for the average person to imagine that AI is infected with a woke mind virus because people see technology as neutral. You can demonize a person or a group so that no one in your tribe listens to them - that liberal professor, for example. It is harder to do that with a programming language.
For these reasons, I am optimistic about how ChatGPT can help construct a shared reality and bring us closer as a nation.