The Future of Truth in the Age of AI

If you haven't been following the news lately, here's something you should know...

Since January 2025:

• Over 8,000 government web pages have been removed or modified.
• The Centers for Disease Control took down more than 3,000 pages.
• The Census Bureau removed about 3,000 pages of research materials.
• The Department of Justice deleted over 180 pages, including state-level crime data and information on anti-LGBTQ hate crimes.
• A national database tracking federal police officer misconduct was eliminated.
• A study showing that white supremacist and far-right violence were the most common forms of domestic terrorism in the United States was removed.

Others in America are also uneasy about the previous administration's efforts to get social media platforms to regulate hate speech. How you feel about either side is a topic for another day, but we need to ask a broader question: How should we consider the integrity of the shared knowledge in the age of AI and corporate influence, and what can we do about it?

Competing for the Crown

Every major tech company is racing to dominate the AI revolution. When OpenAI released ChatGPT, it changed the way we as a society saw the possibilities of AI. At first, it was all fun and games. Then, they were a nonprofit with a mission to ensure that artificial intelligence benefits all of humanity. Now, reports show they've just recently finalized the process to become a for-profit company valued at $500 billion, with Microsoft owning 27%.

I can understand the arguments for why they would make these changes. They're operating in an odd space when it comes to corporate structure, and it takes a lot to compete. However, what we first thought of as a nonprofit entity on a mission for humanity is now struggling with the needs of corporate profits and stakeholders, just like everyone else.

On the other end of the spectrum, you have the richest man in the world, Elon Musk with xAI's Grok. Musk says he's building a "maximum truth-seeking AI", but it has also referred to itself "Mecha Hitler" on one occasion. Musk is also creating an alternative AI generated version of Wikipedia, called Grokipedia, which has raised concerns about privacy, accuracy and political leanings.

Each group is positioning and leveraging themselves to be in the seat of power for the AI revolution. But, what's at stake for all of us? If we don't organize and engage, will we like the future that is handed to us? Let's take a look at the current landscape and break it down from a few angles.

Three Realities We Need to Understand

We can't treat this like we treated social media companies in the early 2000's, or they will run away with the game, too. As consumers, as anyone not in the top 1%, we need to remember that our attention, participation, and purchases will shape this future either way.

So in this high speed, venture capital funded world of innovation at any cost, who can we trust, and how can we establish that trust in a meaningful way? Long story short, we can't trust anyone. We as a society should demand democratization of data, protect and secure human knowledge, and move forward with transparency and checks and balances.

We can split this up into 3 parts:

The Technical Reality: The Synthetic Data Problem

Large language models, like Chat GPT, Claude, or Grok are trained on a set of "training data". At the start, the training data was all the internet content. All natural human context. Now, we're already running up on consuming all that data, and "synthetic data" is a method to add more training data. But, it's been shown that AI models start to get "dumber" when they're trained on content generated by other AI models. Researchers call this "model collapse."

Think of it like making a photocopy of a photocopy of a photocopy. Each generation loses quality, and the details start to disappear. In AI terms, the models lose information about rare but important edge cases—the "tails" of human knowledge—and start producing increasingly similar, less diverse outputs.

Here's the urgent part: human-generated data is finite. Research suggests it will be exhausted by the end of this decade. Once that happens, AI companies will either need to harvest private data, or start feeding AI-generated content back into their models. And, if we aren't protecting our human knowledge, it could be lost in the AI slop, or AI propaganda bot fights in reddit comments.

The race for AI dominance is moving quickly, and we're starting to trust it with our everyday tasks. Will our future human contributions to the LLM training data be recognized for the value it provides? Beyond making deals to protect large companies from Bot Scraping, we don't have much else. And, how do we ensure the proper benefit for the everyday person, not just those with the wealth to protect themselves, like The New York Times or Reddit, who have the capital to go to court in these high profile cases.

What happens when human knowledge becomes a commodity, or a premium service? What happens when the internet becomes too diluted with AI generated content, that we can't tell what's going on? That's already a huge problem.

The Corporate Reality: For Profit Pressures & Loss of Labor

These companies are dealing with massive investments and subsidized business dealings, and we're putting all our eggs in this basket. Industry reports suggest OpenAI has committed to spending over $1.4 trillion over 8 years on infrastructure. And Anthropic, another major AI company founded by former OpenAI employees who left over safety concerns, is now valued at $183 billion (as of September 2025) and has made deals in the billions with Nvidia, Microsoft, Google and Amazon.

The pressure to perform is enormous. When you're trying to raise billions of dollars, investors want returns. They want growth. They want market dominance. Even companies that started with safety-first missions are now navigating the tension between their original ideals and the reality of operating at this scale.

So, where do we (not the 1%) fit in this discussion? We should be fighting for more than jobs in this proposed AI future. Yes, our 40 hour a week job duties may one day become a $99/mo subscription fee. But in my opinion, fighting for "Jobs" is futile and short sighted. We need to pursue equity for all. It shouldn't be up to the ultra-wealthy to decide our future. Here's why.

These companies look at AI as their intellectual property, as something they own. Sure they developed the tech, but our generation's worth of human interaction is what it needs to be useful.

Yes, we should be holding our ground, united under the banner of labor. And, we should be demanding an equitable stake and distribution of the benefits that the future of AI can bring, not just fighting for jobs. Because the magic of AI isn't possible without all of our human knowledge. It belongs to all of us.

The Political Reality: When Information Disappears

Here's where it all comes together. When government agencies remove content—whether it's one party deleting 8,000 pages or the other pressuring platforms to censor—that information becomes harder to find. And when it's harder to find, it's less likely to be in the training data for future AI models.

So in five years, when someone asks an AI about crime statistics, or environmental justice research, or studies on domestic terrorism, the model might give an incomplete answer because that data was systematically removed from the public record. And the user won't even know they're getting an incomplete picture.

This can happen from any direction. It's not about left versus right. It's about the precedent of information being manipulated or disappeared based on who's in power at the moment. And once that information is out of the training data, it creates blind spots that compound over time. Allowing for whomever is in power to more easily control the narrative.

How Should We Respond

What's Really at Stake

The future of truth itself is at stake. Or at least what humans believe is true. With the advancements in AI surveillance tech, government contracts, and an increasingly connected system fueled by private data brokers combining all of our data into profiles that can track and predict our behavior, we're building infrastructure ripe for unprecedented control.

The way we respond now will shape the future for generations to come. If we don't look past our differences, and instead choose to unite under the broader banner of "labor", or "consumer", we will continue to lose in comparison to the ultra-wealthy and powerful. It's not inevitable if we organize.

Focus on Equity, Not Jobs

We've been focused on keeping our jobs, but is that the future we dreamt of as kids after reading or watching our favorite sci-fi film? This feels like a prequel to Star Wars, and the empire is rising. The future of instant knowledge and automation seems inevitable, but how it benefits who is not set in stone. If these technologies can truly revolutionize every industry, then I want to live in a future of possibilities and freedom for all mankind, not one of control and concentrated power.

Here's the pattern: Whether it's AI companies, social media platforms, or any other tech giant, the story is always the same. We create the value—our data, our content, our attention, our labor. They own the infrastructure. But the give and take is tilted much in their favor. I'd argue that the stakes are too high to allow a few to control the outcomes of generations to come. The result of this generation's actions are currently shaping the future of humanity.

The way we fight for a better future is to fight for these tools to be democratically managed, not by government and corporations, but by all of us. We need awareness and consensus to proceed. We need to ensure the development of these tools is focused on benefiting everyday humans, not just fueling corporate interests.

Don't let the powers that be control the narrative. Our discourse and debate should be about a future that includes all of us as stakeholders and shareholders, because LLMs wouldn't be possible without all of our years of content. It's not something some company owns, it's something that all of us make possible. We all deserve to build this future together, not just to get scraps.

The Bottom Line

They're Making Choices About Our Future Without Us

Here's what's really happening: A handful of companies are racing to build AI systems that will shape what future generations can know, using the collective knowledge we all created, funded by investors who want returns measured in months, not generations. This is a race to monetize knowledge even further.

Think about it: A nurse who spent 15 years posting about patient care on Reddit. A mechanic who answered thousands of questions on forums. A teacher who shared lesson plans online. They created knowledge that's now training AI and trapped behind a subscription model.

Did the nurse get paid for that? No. Do they own any of the AI that learned from them? No. Will they benefit from the billions made from their expertise? No. But the companies that scraped that data? They're getting rich. That's not innovation. That's extraction.

We must demand that democratic values shape our collective knowledge base, rather than be determined by whoever has the most money and the loudest voice. This is the only way to protect the future and integrity of what we know to be true.

What Happens Next Is Up to All of Us

All of human knowledge—what we pass down to future generations, what AI systems learn from, what truth even means—shouldn't be controlled by a handful of CEOs or shifting political winds. It should be something we all have a say in.

You don't need to understand transformer models or synthetic data generation. You need to understand power—who has it, who wants more, and what they're doing to get it. When you see headlines about AI breakthroughs, ask: Who benefits? What's not being said? What data made this possible?

But the only way that happens is if we actually show up to have that say. This is going to require ongoing attention, organization and pressure. Whether you're a business leader making product decisions, a parent protecting your child's future, or a developer building open source products to help co-ops compete in capitalism, we need everyone's input.

Five years ago, most of us hadn't heard of OpenAI or ChatGPT. Five years from now, the decisions being made right now will have compounded into permanent patterns. The question isn't whether AI will shape our future—it already is. The question is whether everyday people will have any say in how. Let's work together to protect the truth, fight for real equity, and build the future we dreamt of as kids in the age of technology. Remember, together we win. We the people. Not we the corporations, or we the Republicans, or we the Democrats.