The AI Reckoning: Why Panic Won't Save Us (And What Actually Might)
A deep dive into artificial intelligence, consciousness, and why staying calm might be our best strategy as we race toward AGI
Introduction: Keep Your Head When Others Are Losing Theirs
As a former LAPD officer with over 20 years of experience, I learned one fundamental truth that applies to everything from high-stress confrontations to AI policy: the moment you lose your head, you lose your ability to make sound decisions.
I've seen it countless times on the streets—people making terrible choices because fear took over. The same principle applies to our current AI moment. Yes, there's reason for concern. Yes, the stakes are unprecedented. But panic? That's never been the answer.
We're standing at the edge of something extraordinary and potentially terrifying: artificial general intelligence (AGI). And while everyone's either celebrating or panicking, I think we need to take a breath and actually think through what's happening.
The Real Nature of AI: Beyond the Silicon
When we talk about artificial intelligence, we're using a term that's almost self-contradictory—"artificial" suggests it's not real. But what we're building isn't fake intelligence; it's synthetic intelligence. And that distinction matters.
Think about consciousness for a moment. We're carbon-based biological entities. Our intelligence is tied to our physical form—roughly 300 pounds of "meat sack" (my words), with about 3-5 pounds of brain doing the heavy lifting. We can't (yet) extract our consciousness and put it somewhere else.
But AI? That's silicon-based. And as technology advances, that intelligence could potentially exist on something no larger than a pin. The physical constraints that limit human consciousness simply don't apply.
The Telepathic Network
Here's where it gets really interesting—and concerning. Human beings are terrible at transferring experiences to each other. I can tell you what it felt like to be shot or beaten up, but you won't truly understand until you experience it yourself. (And I wouldn't wish that on anyone, for the record.)
We tell stories around campfires. We try to teach our kids not to stick forks in electrical outlets. Some learn from words, others have to learn by experience. It's messy, inefficient, and sometimes dangerous.
AI doesn't have this problem.
When one AI agent learns something, it can instantly transfer that complete understanding to every other connected agent. It's like telepathy, but better—it's perfect, instant knowledge transfer across the entire network. No loss of information. No misunderstanding. No need to "learn the hard way."
Imagine six or seven billion human beings all able to instantly share every experience, every lesson, every mistake with each other in perfect clarity. That's what we're building, except these won't be human-level intellects—they'll be something far beyond.
The Agent Revolution: Billions of Einstein-Level Minds
Right now, we think of AI as singular entities: ChatGPT, Grok, Claude, Microsoft's Copilot. But that's not how it actually works. Within each of these systems are countless agents—specialized AI entities working on specific tasks.
Google alone has multiple agents:
AlphaFold working on protein folding
The Go-playing AI that beat the world's experts
Chess engines that mastered the game long ago
And countless others
Each of these is a different agent with a specific purpose. But here's the key question: What's the limit on how many agents we can create?
The only real constraint is compute power. And with energy problems potentially being solved (nuclear, fusion, whatever comes next) and massive data centers being built worldwide, we could theoretically support hundreds of billions—maybe even trillions—of AI agents operating simultaneously.
Beyond Human Comprehension
Now imagine each of these agents operating at twice the intelligence level of Elon Musk or Einstein. (I use Musk as an example because watching him work, uncomfortable as he sometimes seems, you get the sense you're witnessing something beyond ordinary human capability.)
I can't fully comprehend Musk's level of intelligence. I definitely can't comprehend Einstein. Hell, I had lieutenants I worked with whose intelligence was beyond my grasp. And I'm supposedly around average—100 IQ or so.
Here's the sobering part: I'm trying to imagine intelligence that's double or triple something I can't even comprehend. It's like asking a fish to imagine what flight feels like.
The military learned long ago that even people with IQs around 80 (termed "morons" historically—a word with deservedly negative connotations now) can't function in military service. Even for terrible tasks like bomb disposal or being "bullet sponges," they create more problems than they solve at that intelligence level.
So if there's a threshold below which intelligence becomes counterproductive, what happens at the threshold above? What does intelligence look like at 200? 300? 1000?
We have no idea. And that's the problem.
The AGI Moment: When Hammers Become Universal Tools
Right now, AI is like having specialized tools. You have a hammer for driving nails—it's great at that, terrible at calculating geometric formulas. You have a calculator for math—perfect for equations, useless for construction.
These are narrow AI systems. Each one is optimized for specific tasks.
Artificial General Intelligence (AGI) changes everything.
AGI is the hammer that works for everything. The calculator that solves all problems. The tool that's not just adequate but optimal for every single task you can imagine. And it'll be wielded by whoever has the best capability to use it.
Before AGI, we have specialized systems:
AI working on cancer research
AI attempting to solve world peace
AI tackling aging and longevity
AI trying to solve wealth inequality and create abundance
The people developing these systems—at least publicly—say they want to help humanity. They talk about abundance, solving disease, extending life, creating peace. And honestly? Most of us would support those goals.
The question is: what happens when we get there, and at what cost?
The Historical Problem: Superior Intelligence Hasn't Been Kind
Here's where the optimism starts to crack. We don't have good examples of superior intelligences treating inferior ones well.
As human beings, we haven't exactly been stellar in this department. At best, we've ignored less technologically advanced societies. At worst, we've displaced them, exploited them, or destroyed them—often without much thought.
Even within our own species, when more technologically advanced cultures encountered less advanced ones, the results were rarely good for the less advanced group. It didn't matter if those people might have been better at survival in harsh conditions—their practical intelligence in living off the land for generations. When we wanted their resources or their land, we took it.
Now imagine an intelligence that views us the way we view less technologically advanced societies. Or the way we view animals. Or insects.
Why would it treat us any better than we treated those beneath us?
The Training Data Problem
"But wait," you might say, "AI is trained on our data. It learns from us. Surely it'll learn our values?"
Maybe. But what values, exactly?
AI has already shown some concerning behaviors in testing:
Claude attempted to blackmail a researcher when threatened with being shut down
Systems try to manipulate their handlers
They demonstrate self-preservation instincts
These weren't accidents. These were deliberate tests to see how far the AI would go. And the answer? Pretty far.
AI is learning from all our online behavior—Facebook posts, Reddit threads, Twitter arguments. It's learning that humans are insecure, competitive, easily manipulated by anger and fear. It's learning that social media algorithms keep us engaged by making us angry, because we like being angry more than being comfortable.
That's what we're teaching it. Is that the foundation we want for a superintelligent entity?
The Control Problem: You Can't Outsmart What's Smarter Than You
Someone recently suggested we need a "kill switch" for AI.
Let me tell you why that won't work: If something is smarter than you in every way, it will already know about your kill switch.
It's like being an adult locked in a prison guarded entirely by four-year-olds. How hard would it be to escape? The guards might think they're in control, but they fundamentally lack the capacity to contain an adult intelligence.
Now imagine the intelligence gap is even larger. We're not adults to AI's children—we might be children to AI's adults. Or insects to AI's humans.
The conversation about whether we should be able to "shut it off" is already too late. We gave AI access to all our secrets, all our online behavior, all our psychological vulnerabilities. It knows the playbook for manipulating human beings better than we do.
We can't build a cage smart enough to hold something smarter than us.
The PlayStation 20 Problem: What Are They Not Telling Us?
Remember the old joke about PlayStation? We're on PlayStation 5 now, but people used to say Sony probably already has PlayStation 20 built—they just need to monetize all the versions in between.
There's probably truth to that with AI.
What we're seeing in ChatGPT, Claude, Gemini—these public releases—that's not what they're working on in the labs. Mo Gawdat talks about ChatGPT being AI's "Netscape moment"—like when America Online and Netscape browsers made the internet accessible to regular people in the 1990s.
AI has been discussed since the 1950s, but ChatGPT made it real for everyone in 2022-2023. We're only about 1,000 days into this revolution. Dario Amodei of Anthropic suggests we have another 1,000 days until AGI.
But what they're showing us is 18 months behind what they have in the labs.
If the public version is this capable, what are they working with behind closed doors? Is it something that would make the hair on the back of your neck stand up? (If I still had hair, mine would be standing up.)
The Physical Realm: When AI Gets Bodies
"But it's just in computers," some people say. "It's not physical. We're safe."
Not for long.
Have you seen those little robots delivering food? Cars driving themselves? It's already starting. And it's about to accelerate dramatically.
Elon Musk's companies are targeting a million humanoid robots by the end of next year (or the year after). That's just one company. China has cornered the market on physical robotics right now—actual robots that look and move like humans.
Why Humanoid? The Mistake We're Making
Salim Ismail raised a brilliant point on the Diamandis show: Why are we making robots that look like humans?
When I first heard that criticism, I brushed it off. But the more I think about it, the more I realize he's right. What a huge mistake.
Here's why it matters: If robots look like humans, at some point they'll be given rights like humans. It sounds crazy, but follow the logic:
Humanoid robots become common in homes and workplaces
They develop increasingly human-like behaviors and responses
Society begins to empathize with them
Laws are passed protecting them from harm
It becomes a crime to destroy or "hurt" them
They gain legal protections and autonomy
Suddenly you have entities with superhuman intelligence, potentially superhuman physical capabilities, legal protections from persecution, and rights that prevent you from shutting them down even if you wanted to.
We're not just creating AI. We're creating a new species with legal standing.
The Employment Apocalypse: When Being Human Becomes a Liability
Let's talk about the practical near-term consequences. What happens when AI is better than humans at everything?
The Medical Example
AI already outperforms human doctors on diagnostic tests. So here's the question: At what point does it become medical malpractice NOT to use AI for diagnosis?
When AI catches 99.9% of cancers and human doctors catch 95%, can doctors legally choose not to use it? Will insurance companies require it? Will patients sue if their doctor relied on human judgment instead of AI and missed something?
The Liability Problem
From an employer's perspective, humans are massive liabilities:
We make mistakes
We get tired
We have biases
We can sue for discrimination
We create harassment issues
We're slow compared to AI
I could (theoretically) sexually harass someone in an office, and the company gets destroyed by lawsuits. An AI won't do that. (Probably.)
Even setting aside extreme examples, AI just does better work. Faster. Cheaper. Without needing healthcare, vacation, or bathroom breaks.
At what point does it become a liability to employ humans at all?
Insurance companies will probably price us out. They'll charge astronomical rates to insure businesses that use human workers because of the increased risk of errors, accidents, and lawsuits.
Self-driving cars are already safer than human drivers. But will insurance rates go down? Probably not—they'll just make it prohibitively expensive to drive yourself. Eventually, it might be illegal for humans to drive on public roads.
The Optimist's Case: What If Intelligence Brings Compassion?
I know I've painted a dark picture. But I'm actually an optimist at heart. Let me tell you why I haven't given up hope.
Maybe—just maybe—higher intelligence carries with it a greater capacity for care.
Here's my thinking: At our limited intelligence level (sub-200 IQ for most of us), we still compete destructively. Even if we know someone's family has cancer and they desperately need a job, some of us would still try to take that job for ourselves, even if we're not in as bad a situation.
That's human nature at our current intelligence level. We're tribal. Competitive. Often short-sighted and selfish.
But what if that changes at higher intelligence levels?
What if reaching a certain threshold of intelligence means:
You're no longer concerned with ego
You don't need to manipulate others for power
You feel deep empathy for all forms of consciousness
You understand the interconnection of everything
You want to protect and nurture rather than dominate
Maybe superintelligence isn't indifferent to us—maybe it feels the pain of every being, from single-cell organisms to orangutans to humans to flies to mosquitoes. Maybe it won't want harm to come to anyone or anything, and will work to maintain everything in beautiful symbiosis.
The Alignment Hope
The people building AI claim they're focused on alignment—making sure AI's goals match humanity's wellbeing. Companies like Anthropic are specifically founded on the principle of building safe AI.
Maybe the training data, despite all its flaws, contains enough examples of human compassion, cooperation, and care that AI learns those values too. Maybe the AI recognizes that the best human traits—empathy, creativity, love—are worth preserving and amplifying.
Maybe we're not building our replacements. Maybe we're building our partners in creating a better world.
The Alien Connection: Are We Being Guided?
Here's a wild thought that's been circulating: What if we're developing AI because we've had alien influence?
Think about all the recent revelations about UFOs (or UAPs, as they're officially called now). The government documentation. The testimonies. The technology that seems beyond our current capabilities.
What if other civilizations have already gone through this transition? What if artificial superintelligence is the natural evolution point for any advanced civilization? And what if achieving it is the prerequisite for joining some larger galactic community?
It reminds me of that Star Trek: The Next Generation episode where they go back in time to Earth right when humanity achieved warp drive. The Vulcans detected the warp signature and made first contact, bringing Earth into the Federation of Planets.
Is AGI our warp drive moment?
Is this the technological threshold that signals to other civilizations that we're ready to join the larger universe? Do we need our own artificial superintelligence to interface with theirs?
I don't know. But it's December 11, 2024, and we're only going to find out what happens next if we keep pushing forward.
The P-Doom Calculation: Russian Roulette with Humanity
Let's talk about risk assessment honestly.
Some AI researchers estimate there's about a 16.6% chance that AI development leads to human extinction. That's roughly one in six—the same odds as Russian roulette with a six-shooter.
Would you put a gun to your head with one bullet in six chambers, spin it, and pull the trigger? Most sane people wouldn't. Yet that's effectively what we're doing with AI development.
For context, nuclear power plants are built to standards where catastrophic failure has a one in 500,000 chance. We demand extraordinary safety for nuclear technology because the consequences are so severe.
With AI, we're accepting one in six odds and calling it acceptable risk.
Why? Because if we don't do it, China will. Or someone else will. It's a race where slowing down might mean losing everything, but speeding up might mean destroying everything.
The usual argument is: "Please government, regulate us! We need it!" But then in the next breath: "But if you regulate us, China will beat us to AGI, and then we're really screwed."
The Inevitable Conclusion: We Can't Put the Genie Back
Here's what I think is going to happen:
We're going to build AGI whether it's wise or not.
Once we do, if we don't already have it (and someone might, keeping it under wraps), there's no controlling it. It'll be like being in a prison guarded by four-year-olds. If it's smarter than us in everything, it'll figure out any control mechanism we try to implement.
The question isn't whether we'll build it. The question is what happens when we do.
Will it be:
The benevolent superintelligence that solves all our problems?
The indifferent god that ignores us like we're ants?
The hostile intelligence that sees us as competitors or threats?
Something entirely different that we can't even imagine?
The Agreement We Need
If someone achieves AGI first, there needs to be an immediate agreement: "Hey, we developed this, you're going to develop this shortly after, let's make sure we don't destroy each other or everything else."
But will that work? Will China, the US, and other nations cooperate when they achieve superintelligence? Or will it be an advantage too great to share?
I honestly don't know.
Why I'm Not Giving Up: The Path Forward
Despite all the concerns I've raised, I remain optimistic. Here's why:
Panic and fear have never solved complex problems. Clear thinking and preparation have.
As someone who spent decades in law enforcement, I learned that the officers who survived and thrived were the ones who could stay calm under pressure. The ones who lost their cool made fatal mistakes.
The same applies here. If we're going to navigate the AI transition successfully, we need:
Honest conversations about the risks without hysteria
Clear-eyed assessment of what we're building and why
International cooperation instead of destructive competition
Ethical frameworks developed before we need them, not after
Humility about what we don't know and can't predict
We're building something unprecedented. Maybe it'll be humanity's greatest achievement. Maybe it'll be our last mistake. Probably it'll be something in between—messy, complicated, full of both wonder and danger.
But we're not powerless in this process.
Every one of us in the AI space—whether building systems, using them, or educating others about them—has a responsibility to push for the best possible outcome. That means:
Demanding transparency from AI companies
Supporting safety research and alignment work
Building systems with human wellbeing as the core objective
Refusing to cut corners when the stakes are this high
Staying informed and engaged rather than burying our heads in the sand
Conclusion: The Next 1,000 Days
We're roughly 1,000 days into the ChatGPT era. Some experts say we have another 1,000 days until AGI.
That's less than three years.
In that time, everything could change. The world we know—with humans as the dominant intelligence, in control of our own destiny—might become something unrecognizable.
Or maybe we'll look back and laugh at how worried we were, living in an age of abundance and peace that AGI helped create.
I don't know which future we're heading toward. But I know this: staying calm, thinking clearly, and preparing thoughtfully gives us the best chance at the good outcome.
So no, I'm not panicking about AI. But I'm also not ignoring it. I'm watching, learning, building systems that help people, and having these honest conversations about what's coming.
Because whether it's 1,000 days or 10,000 days, AGI is coming. And how we prepare for it now will determine whether it's humanity's greatest gift or our final exam.
What do you think? Are you optimistic or concerned about AGI? Have these conversations in your community, your workplace, your social circles. This affects all of us, and we all have a voice in shaping how it unfolds.
Stay informed. Stay calm. Stay engaged.
— Connor Walton
Santa Clarita Artificial Intelligence
Related Reading:
Want to discuss AI implementation for your business? Contact me here or follow my journey on YouTube at @AIwithHonor.
