Month: August 2025

I Really Need To Go Back To Seoul Eventually

by Shelt Garner
@sheltgarner

Ho hum.

For some reason, I find myself thinking of Seoul AGAIN. I keep thinking about all the adventures I had while I was in Asia and how nice it would be to go back and have people actually…care. That was probably the biggest difference between now and then — back in my Seoul days, people actually gave a shit about me.

Now…lulz.

I am well aware that if I went back that it would be a very harsh reality. Everyone I knew from way back when are long gone. It probably would seem very, very boring. There might be a few Koreans who remember me, but, I don’t know, I just would have to manage my expectations.

And, what’s more, I’m not going back to Asia anytime soon. It could be years and I’ll be even older than I am now. It’s all just kind of sad. I could be dating a robot by the time I have the funds to go back to Asia.

Sigh.

Will The Little Gold Men Question Issue Be My Summer Lull Hijinks For This Year?

by Shelt Garner
@sheltgarner

As I’ve written before, every year about this time of year, a small something happens that puts a twinkle in my life and a spring in my step. Since I lead a very, very, very dull life, it doesn’t have to be much — as me writing this blog post about this very mundane thing is a testament.

Anyway, as I have mentioned before, I sent in a few questions to the Little Gold Men podcast (they kept begging for listener questions) and I have a tiny little hunch that maybe they do a search of my name and found this blog. I don’t know, but…maybe?

Regardless, at least I have something to think about, I guess. And I’m not going to tell you I’m, like, waiting with baited breath for them to potentially read my question on the podcast. There’s a pretty good chance I’ll miss it if they do read one of my questions because I only listen to the podcast in a half-assed kind of way.

Don’t get me wrong, it’s a great pod, it’s just I’m not really obsessed with any pods these days since Pod Save America kind of jumped the shark about a year ago. And this reminds me of my personal belief that before AI takes over everything that there’s a market for a Gawker-like pod that would be really snarky and come out five days a week.

Anyway. I find it amusing that Little Gold Men thought a very, very niche movie like Fire Island would…be a hit? Wow was that a niche movie and there simply no way that movie was going to do anything but arthouse numbers. And, yet, there I was, listening to the Little Gold Men people effuse about the movie as if it was Raging Bull or something.

But I have to be honest with myself — I listen to the Little Gold Men podcast because of it’s snooty point of view and how insiderish it is. That’s the whole point. That’s the fun of it all, listening to what “woke” (wink) snobs think of movies. Lol.

I Hope This Works Out

By Shelt Garner
@sheltgarner

The key issue about the novel I’m working on now is it’s very character driven. There’s no mystery to solve, it’s more of a “boy meets girl, boy loses girl, boy gets girl” type of story. I’m hoping that if the story is simple enough that I can actually finish this damn thing, instead of spinning my wheels.

Now, obvious, the case could be made if I had, like, friends and stuff, that maybe I wouldn’t have spent so many years on a thriller that went no where in various forms. But…I’m ok with what happened.

I really think this novel is going to turn out ok. I’m really going to steer clear of having AI write too much of the novel during the development process because I don’t want to get discouraged because my writing just isn’t as good in comparison.

Finding My Novel: A Writer’s Journey to Creative Momentum

After years of false starts and abandoned manuscripts, I think I’ve finally cracked the code. Not the secret to writing the Great American Novel, mind you—just the secret to writing a novel. And sometimes, that’s exactly what you need.

The Ambition Trap

Looking back, I can see where I went wrong before. Every time I sat down to write, I was trying to craft something profound, something that would change literature forever. I’d create these sprawling, complex narratives with intricate world-building and dozens of characters, each with their own detailed backstories and motivations.

The problem? I’d burn out before I even reached the middle of Act One.

This time feels different. I’ve stumbled across an idea that excites me—not because it’s going to revolutionize fiction, but because it’s something I can actually finish. There’s something liberating about embracing a concept that’s focused, manageable, and most importantly, writeable at speed.

The AI Dilemma

I’ve had to learn some hard lessons about artificial intelligence along the way. Don’t get me wrong—AI is an incredible tool for certain tasks. Rewriting blog posts like this one? Perfect. Getting unstuck on a particularly stubborn paragraph? Helpful. But when it comes to the heart of creative work, I’ve discovered that AI can be more hindrance than help.

There’s nothing quite like the deflating feeling of watching AI generate a first draft that’s objectively better than anything you could produce as a human writer. It’s efficient, polished, and technically proficient in ways that can make your own rough, imperfect human voice feel inadequate by comparison.

But here’s what I’ve realized: that technical perfection isn’t what makes a story worth telling. The messy, flawed, uniquely human perspective—that’s where the magic happens. That’s what readers connect with, even if the prose isn’t as smooth as what a machine might produce.

The Path Forward

I have an outline now. Nothing fancy, but it’s solid and it’s mine. My plan is to flesh it out methodically, then dive into the actual writing. Though knowing myself, I might get impatient and just start writing, letting the story evolve organically and adjusting the outline as I go.

Both approaches have their merits. The disciplined, outline-first method provides structure and prevents those dreaded “now what?” moments. But there’s also something to be said for the discovery that happens when you just put words on the page and see where they take you.

The Real Victory

What I’m chasing isn’t literary acclaim or critical recognition—it’s that moment when I can type “The End” and feel the deep satisfaction of having completed something truly substantial. There’s a unique pride that comes with finishing a novel, regardless of its ultimate quality or commercial success. It’s the pride of having sustained focus, creativity, and determination long enough to build an entire world from nothing but words.

The creative momentum is building. For the first time in years, I feel like I have a story that wants to be told and the practical framework to tell it. Whether I’ll stick to the outline or let inspiration guide me, I’m ready to find out.

Wish me luck. I have a feeling I’m going to need it—and more importantly, I’m finally ready to earn it.

The Perceptual Shift: How Ubiquitous LLMs Will Restructure Information Ecosystems

The proliferation of powerful, personal Large Language Models (LLMs) integrated into consumer devices represents a pending technological shift with profound implications. Beyond enhancing user convenience, this development is poised to fundamentally restructure the mechanisms of information gathering and dissemination, particularly within the domain of journalism and public awareness. The integration of these LLMs—referred to here as Navis—into personal smartphones will transform each device into an autonomous data-gathering node, creating both unprecedented opportunities and complex challenges for our information ecosystems.

The Emergence of the “Datasmog”

Consider a significant public event, such as a natural disaster or a large-scale civil demonstration. In a future where LLM-enabled devices are ubiquitous, any individual present can become a source of high-fidelity data. When a device is directed toward an event, its Navi would initiate an autonomous process far exceeding simple video recording. This process includes:

  • Multi-Modal Analysis: Real-time analysis of visual and auditory data to identify objects, classify sounds (e.g., differentiating between types of explosions), and track movement.
  • Metadata Correlation: The capture and integration of rich metadata, including precise geospatial coordinates, timestamps, and atmospheric data.
  • Structured Logging: The generation of a coherent, time-stamped log of AI-perceived events, creating a structured narrative from chaotic sensory input.

The collective output from millions of such devices would generate a “datasmog”: a dense, overwhelming, and continuous flood of information. This fundamentally alters the landscape from one of information scarcity to one of extreme abundance.

The Evolving Role of the Journalist

This paradigm shift necessitates a re-evaluation of the journalist’s role. In the initial phases of a breaking story, the primary gathering of facts would be largely automated. The human journalist’s function would transition from direct observation to sophisticated synthesis. Expertise will shift from primary data collection to the skilled querying of “Meta-LLM” aggregators—higher-order AI systems designed to ingest the entire datasmog, verify sources, and construct coherent event summaries. The news cycle would compress from hours to seconds, driven by AI-curated data streams.

The Commercialization of Perception: Emergent Business Models

Such a vast resource of raw data presents significant commercial opportunities. A new industry of “Perception Refineries” would likely emerge, functioning not as traditional news outlets but as platforms for monetizing verified reality. The business model would be a two-sided marketplace:

  • Supply-Side Dynamics: The establishment of real-time data markets, where individuals are compensated via micropayments for providing valuable data streams. The user’s Navi could autonomously negotiate payment based on the quality, exclusivity, and relevance of its sensory feed.
  • Demand-Side Dynamics: Monetization would occur through tiered Software-as-a-Service (SaaS) models. Clients, ranging from news organizations and insurance firms to government agencies, would subscribe for different levels of access—from curated video highlights to queryable metadata and even generative AI tools capable of creating virtual, navigable 3D models of an event from the aggregated data.

The “Rashomon Effect” and the Fragmentation of Objective Truth

A significant consequence of this model is the operationalization of the “Rashomon Effect,” where multiple, often contradictory, but equally valid subjective viewpoints can be accessed simultaneously. Users could request a synthesis of an event from the perspectives of different participants, which their own Navi could compile and analyze. While this could foster a more nuanced understanding of complex events, it also risks eroding the concept of a single, objective truth, replacing it with a marketplace of competing, verifiable perspectives.

Conclusion: Navigating the New Information Landscape

The advent of the LLM-driven datasmog represents a pivotal moment in the history of information. It promises a future of unparalleled transparency and immediacy, particularly in public safety and civic awareness. However, it also introduces systemic challenges. The commercialization of raw human perception raises profound ethical questions. Furthermore, this new technological layer introduces new questions regarding cognitive autonomy and the intrinsic value of individual, unverified human experience in a world where authenticated data is a commodity. The primary challenge for society will be to develop the ethical frameworks and critical thinking skills necessary to navigate this complex and data-saturated future.

When AI Witnesses History: How the LLM Datasmog Will Transform Breaking News

7:43 AM, San Francisco, the day after tomorrow

The ground shakes. Not the gentle rolling of a typical California tremor, but something violent and sustained. In that instant, ten thousand smartphone LLMs across the Bay Area simultaneously shift into high alert mode.

This is how breaking news will work in the age of ubiquitous AI—not through human reporters racing to the scene, but through an invisible datasmog of AI witnesses that see everything, process everything, and instantly connect the dots across an entire city.

The First Ten Seconds

7:43:15 AM: Sarah Chen’s iPhone AI detects the seismic signature through accelerometer data while she’s having coffee in SOMA. It immediately begins recording video through her camera, cataloging the swaying buildings and her startled reaction.

7:43:18 AM: Across the city, 847 other smartphone AIs register similar patterns. They automatically begin cross-referencing: intensity, duration, epicenter triangulation. Without any human intervention, they’re already building a real-time earthquake map.

7:43:22 AM: The collective AI network determines this isn’t routine. Severity indicators trigger the premium breaking news protocol. Thousands of personal AIs simultaneously ping the broader network: “Major seismic event detected. Bay Area. Magnitude 6.8+ estimated. Live data available.”

The Information Market Ignites

7:44 AM: News organizations’ AI anchors around the world receive the alerts. CNN’s AI anchor immediately starts bidding for access to the citizen AI network. So does BBC, Reuters, and a hundred smaller outlets.

7:45 AM: Premium surge pricing kicks in. Sarah’s AI, which detected some of the strongest shaking, receives seventeen bid requests in ninety seconds. NBC’s AI anchor offers $127 for exclusive ten-minute access to her AI’s earthquake data and local observations.

Meanwhile, across millions of smartphones, people’s personal AI anchors are already providing real-time briefings: “Major earthquake just hit San Francisco. I’m accessing live data from 800+ AI witnesses in the area. Magnitude estimated at 6.9. No major structural collapses detected yet, but I’m monitoring. Would you like me to connect you with a seismologist twin for context, or pay premium for live access to Dr. Martinez who’s currently at USGS tracking this event?”

The Human Premium

7:47 AM: Dr. Elena Martinez, the USGS seismologist on duty, suddenly finds herself in the highest-demand breaking news auction she’s ever experienced. Her live expertise is worth $89 per minute to news anchors and individual consumers alike.

But here’s what’s remarkable: she doesn’t have to manage this herself. Her representation service automatically handles the auction, booking her for twelve-minute live interview slots at premium rates while she focuses on the actual emergency response.

Meanwhile, the AI twins of earthquake experts are getting overwhelmed with requests, but they’re offering context and analysis at standard rates to anyone who can’t afford the live human premium.

The Distributed Investigation

7:52 AM: The real power of the LLM datasmog becomes clear. Individual smartphone AIs aren’t just passive observers—they’re actively investigating:

  • Pattern Recognition: AIs near the Financial District notice several building evacuation alarms triggered simultaneously, suggesting potential structural damage
  • Crowd Analysis: AIs monitoring social media detect panic patterns in specific neighborhoods, identifying areas needing emergency response
  • Infrastructure Assessment: AIs with access to traffic data notice BART system shutdowns and highway damage, building a real-time map of transportation impacts

8:05 AM: A comprehensive picture emerges that no single human reporter could have assembled. The collective AI network has mapped damage patterns, identified the most affected areas, tracked emergency response deployment, and even started predicting aftershock probabilities by consulting expert twins in real-time.

The Revenue Reality

By 8:30 AM, the breaking news economy has generated serious money:

  • Citizen AI owners who were near the epicenter earned $50-300 each for their AIs’ firsthand data
  • Expert representation services earned thousands from live human seismologist interviews
  • News organizations paid premium rates but delivered unprecedented coverage depth to their audiences
  • Platform companies took their cut from every transaction in the citizen AI marketplace

What This Changes

This isn’t just faster breaking news—it’s fundamentally different breaking news. Instead of waiting for human reporters to arrive on scene, we get instant, comprehensive coverage from an army of AI witnesses that were already there.

The economic incentives create better information, too. Citizens get paid when their AIs contribute valuable breaking news data, so there’s financial motivation for people to keep their phones charged and their AIs updated with good local knowledge.

And the expert twin economy provides instant context. Instead of waiting hours for expert commentary, every breaking news event immediately has analysis available from AI twins of relevant specialists—seismologists for earthquakes, aviation experts for plane crashes, geopolitical analysts for international incidents.

The Datasmog Advantage

The real breakthrough is the collective intelligence. No single AI is smart enough to understand a complex breaking news event, but thousands of them working together—sharing data, cross-referencing patterns, accessing expert knowledge—can build comprehensive understanding in minutes.

It’s like having a newsroom with ten thousand reporters who never sleep, never miss details, and can instantly access any expert in the world. The datasmog doesn’t just witness events—it processes them.

The Breaking News Economy

This creates a completely new economic model around information scarcity. Instead of advertising-supported content that’s free but generic, we get surge-priced premium information that’s expensive but precisely targeted to what you need to know, when you need to know it.

Your personal AI anchor becomes worth its subscription cost precisely during breaking news moments, when its ability to navigate the expert marketplace and process the citizen AI datasmog becomes most valuable.

The Dark Side

Of course, this same system that can rapidly process an earthquake can also rapidly spread misinformation if the AI witnesses are compromised or if bad actors game the citizen network. The premium placed on being “first” in breaking news could create incentives for AIs to jump to conclusions.

But the economic incentives actually favor accuracy—AIs that consistently provide bad breaking news data will get lower bids over time, while those with reliable track records command premium rates.

The Future Is Witnessing

We’re moving toward a world where every major event will be instantly witnessed, processed, and contextualized by a distributed network of AI observers. Not just recorded—actively analyzed by thousands of artificial minds working together to understand what’s happening.

The earthquake was just the beginning. Tomorrow it might be a terrorist attack, a market crash, or a political crisis. But whatever happens, the datasmog will be watching, processing, and immediately connecting you to the expertise you need to understand what it means.

Your personal AI anchor won’t just tell you what happened. It will help you understand what happens next.

In the premium breaking news economy, attention isn’t just currency—it’s the moment when artificial intelligence proves its worth.

The Expert Twin Economy: How AI Will Make Every Genius Accessible for $20 a Month

Forget chatbots. The real revolution in AI isn’t about building smarter assistants—it’s about creating a world where your personal AI can interview any expert, living or dead, at any time, for the cost of a Netflix subscription.

The Death of Static Media

We’re rapidly approaching a world where everyone has an LLM as firmware in their smartphone. When that happens, traditional media consumption dies. Why listen to a three-hour Joe Rogan podcast when your AI can interview that same physicist for 20 minutes, focusing exactly on the quantum computing questions that relate to your work in cryptography?

Your personal AI becomes your anchor—like Anderson Cooper, but one that knows your interests, your knowledge level, and your learning style. When you need expertise, it doesn’t search Google or browse Wikipedia. It conducts interviews.

The Expert Bottleneck Problem

But here’s the obvious issue: experts have day jobs. A leading cardiologist can’t spend eight hours fielding questions from thousands of AIs. A Nobel laureate has research to do, not interviews to give.

This creates the perfect setup for what might become one of the most lucrative new industries: AI Expert Twins.

Creating Your Knowledge Double

Picture this: Dr. Sarah Chen, a leading climate scientist, spends 50-100 hours working with an AI company to create her “knowledge twin.” They conduct extensive interviews, feed in her papers and talks, and refine the AI’s responses until it authentically captures her expertise and communication style.

Then Dr. Chen goes back to her research while her AI twin works for her 24/7, fielding interview requests from personal AIs around the world. Every time someone’s AI wants to understand carbon capture technology or discuss climate tipping points, her twin is available for a virtual consultation.

The economics are beautiful: Dr. Chen earns passive income every time her twin is accessed, while remaining focused on her actual work. Meanwhile, millions of people get affordable access to world-class climate expertise through their personal AIs.

The Subscription Reality

Micropayments sound elegant—pay $2.99 per AI interview—but credit card fees would kill the economics. Instead, we’ll see subscription bundles that make expertise accessible at unprecedented scale:

Basic Expert Access ($20/month): University professors, industry practitioners, working professionals across hundreds of specialties

Premium Tier ($50/month): Nobel laureates, bestselling authors, celebrity chefs, former CEOs

Enterprise Level ($200/month): Ex-presidents, A-list experts, exclusive access to the world’s most sought-after minds

Or vertical bundles: “Science Pack” for $15/month covers researchers across physics, biology, and chemistry. “Business Pack” for $25/month includes MBA professors, successful entrepreneurs, and industry analysts.

The Platform Wars

The companies that build these expert twin platforms are positioning themselves to capture enormous value. They’re not just booking agents—they’re creating scalable AI embodiments of human expertise.

These platforms would handle:

  • Twin Creation: Working with experts to build authentic AI representations
  • Quality Control: Ensuring twins stay current and accurate
  • Discovery: Helping personal AIs find the right expert for any question
  • Revenue Distribution: Managing subscriptions and expert payouts

Think LinkedIn meets MasterClass meets Netflix, but operating at AI speed and scale.

Beyond Individual Experts

The really interesting development will be syndicated AI interviews. Imagine Anderson Cooper’s AI anchor conducting a brilliant interview with a leading epidemiologist about pandemic preparedness. That interview becomes intellectual property that can be licensed to other AI platforms.

Your personal AI might say: “I found an excellent interview that BBC’s AI conducted with that researcher last week. It covered exactly your questions. Should I license it for my Premium tier, or would you prefer I conduct a fresh interview with their twin?”

The best interviewer AIs—those that ask brilliant questions and draw out insights—become content creation engines that can monetize the same expert across millions of individual consumers.

The Democratization of Genius

This isn’t just about convenience—it’s about fundamentally democratizing access to expertise. Today, only Fortune 500 companies can afford consultations with top-tier experts. Tomorrow, anyone’s AI will be able to interview their digital twin for the cost of a monthly subscription.

A student in rural Bangladesh could have their AI interview Nobel laureate economists about development theory. An entrepreneur in Detroit could get product advice from successful Silicon Valley founders. A parent dealing with a sick child could access pediatric specialists through their AI’s interview.

The Incentive Revolution

The subscription model creates fascinating incentives. Instead of experts optimizing their twins for maximum interview volume (which might encourage clickbait responses), they optimize for subscriber retention. The experts whose twins provide the most ongoing value—the ones people keep their subscriptions active to access—earn the most.

This rewards depth, accuracy, and genuine insight over viral moments or controversial takes. The economic incentives align with educational value.

What This Changes

We’re not just talking about better access to information—we’re talking about a fundamental restructuring of how knowledge flows through society. When any personal AI can interview any expert twin at any time, the bottlenecks that have constrained human learning for millennia start to disappear.

The implications are staggering:

  • Education becomes infinitely personalized and accessible
  • Professional development accelerates as workers can interview industry leaders
  • Research speeds up as scientists can quickly consult experts across disciplines
  • Decision-making improves as anyone can access world-class expertise

The Race That’s Coming

The companies that recognize this shift early and invest in building the most compelling expert twin platforms will create some of the most valuable businesses in history. Not because they have the smartest AI, but because they’ll democratize access to human genius at unprecedented scale.

The expert twin economy is coming. The only question is: which platform will you subscribe to when you want your AI to interview Einstein’s digital twin about relativity?

Your personal AI anchor is waiting. And so are the experts.

Your AI Anchor: How the Future of News Will Mirror Television’s Greatest Innovation

We’re overthinking the future of AI-powered news consumption. Everyone’s debating whether we’ll have one personal AI or multiple specialized news AIs competing for our attention. But television already solved this problem decades ago with one of media’s most enduring innovations: the anchor-correspondent model.

The Coming Fragmentation

Picture this near future: everyone has an LLM as firmware in their smartphone. You don’t visit news websites anymore—you just ask your AI “what’s happening today?” The web becomes an API layer that your personal AI navigates on your behalf, pulling information from hundreds of sources and synthesizing it into a personalized briefing.

This creates an existential crisis for news organizations. Their traditional model—getting you to visit their sites, see their ads, engage with their brand—completely breaks down when your AI is extracting their content into anonymous summaries.

The False Choice

The obvious solutions seem to be:

  1. Your personal AI does everything – pulling raw data from news APIs and repackaging it, destroying news organizations’ brand value and economic models
  2. Multiple specialized news AIs compete for your attention – creating a fragmented experience where you’re constantly switching between different AI relationships

Both approaches have fatal flaws. The first commoditizes journalism into raw data feeds. The second creates cognitive chaos—imagine having to build trust and rapport with dozens of different AI personalities throughout your day.

The Anchor Solution

But there’s a third way, and it’s been hiding in plain sight on every evening newscast for the past 70 years.

Your personal AI becomes your anchor—think Anderson Cooper or Lester Holt. It’s the trusted voice that knows you, maintains context across topics, and orchestrates your entire information experience. But when you need specialized expertise, it brings in correspondents.

“Now let’s go to our BBC correspondent for international coverage…”

“For market analysis, I’m bringing in our Bloomberg specialist…”

“Let me patch in our climate correspondent from The Guardian…”

Your anchor AI maintains the primary relationship while specialist AIs from news organizations provide deep expertise within that framework.

Why This Works

The anchor-correspondent model succeeds because it solves multiple problems simultaneously:

For consumers: You maintain one trusted AI relationship that knows your preferences, communication style, and interests. No relationship fragmentation, no switching between personalities. Your anchor provides continuity and context while accessing the best expertise available.

For news organizations: They can charge premium rates for their “correspondent” AI access—potentially more than direct subscriptions since they’re being featured as the expert authority within millions of personal briefings. They maintain brand identity and demonstrate specialized knowledge without having to compete for your primary AI relationship.

For the platform: Your anchor AI becomes incredibly valuable because it’s not just an information service—it’s a cognitive relationship that orchestrates access to the world’s expertise. The switching costs become enormous.

The Editorial Intelligence Layer

Here’s where it gets really interesting. Your anchor doesn’t just patch in correspondents—it editorializes about them. It might say: “Now let’s get the BBC’s perspective, though keep in mind they tend to be more cautious on Middle East coverage” or “Bloomberg is calling this a buying opportunity, but remember they historically skew optimistic on tech stocks.”

Your anchor AI becomes an editorial intelligence layer, helping you understand not just what different sources are saying, but how to interpret their perspectives. It learns your biases and blind spots, knows which sources you trust for which topics, and can provide meta-commentary about the information landscape itself.

The Persona Moat

The anchor model also creates the deepest possible moat. Your anchor AI won’t just know your news preferences—it will develop a personality, inside jokes, ways of explaining things that click with your thinking style. It will become, quite literally, your cognitive companion for navigating reality.

Once that relationship is established, switching to a competitor becomes almost unthinkable. It’s not about features or even accuracy—it’s about cognitive intimacy. Just as viewers develop deep loyalty to their favorite news anchors, people will form profound attachments to their AI anchors.

The New Value Chain

In this model, the value chain looks completely different:

  • Personal AI anchors capture the relationship and orchestration value
  • News organizations become premium correspondent services, monetizing expertise rather than attention
  • Platforms that can create the most trusted, knowledgeable, and personable anchors win the biggest prize in media history

We’re not just talking about better news consumption—we’re talking about a fundamental restructuring of how humans access and process information.

Beyond News

The anchor-correspondent model will likely extend far beyond news. Your AI anchor might bring in specialist AIs for medical advice, legal consultation, financial planning, even relationship counseling. It becomes your cognitive chief of staff, managing access to the world’s expertise while maintaining the continuity of a single, trusted relationship.

The Race That Hasn’t Started

The companies that recognize this shift early—and invest in creating the most compelling AI anchors—will build some of the most valuable platforms in human history. Not because they have the smartest AI, but because they’ll sit at the center of billions of people’s daily decision-making.

The Knowledge Navigator wars are coming. And the winners will be those who understand that the future of information isn’t about building better search engines or chatbots—it’s about becoming someone’s most trusted voice in an increasingly complex world.

Your AI anchor is waiting. The question is: who’s going to create them?

The Coming Knowledge Navigator Wars: Why Your Personal AI Will Be Worth Trillions

We’re obsessing over the wrong question in AI. Everyone’s asking who will build the best chatbot or search engine, but the real prize is something much bigger: becoming your personal Knowledge Navigator—the AI that sits at the center of your entire digital existence.

The End of Destinations

Think about how you consume news today. You visit websites, open apps, scroll through feeds. You’re a tourist hopping between destinations in the attention economy. But what happens when everyone has an LLM as firmware in their smartphone?

Suddenly, you don’t visit news sites—you just ask your AI “what should I know today?” You don’t browse—you converse. The web doesn’t disappear, but it becomes an API layer that your personal AI navigates on your behalf.

This creates a fascinating structural problem for news organizations. The traditional model—getting you to visit their site, see their ads, engage with their brand—completely breaks down when your AI is just extracting and synthesizing information from hundreds of sources into anonymous bullet points.

The Editorial Consultant Future

Here’s where it gets interesting. News organizations can’t compete to be your primary AI—that’s a platform play requiring massive capital and infrastructure. But they can compete to be trusted editorial modules within whatever AI ecosystem wins.

Picture this: when you ask about politics, your AI shifts into “BBC mode”—using their editorial voice, fact-checking standards, and international perspective. Ask about business and it switches to “Wall Street Journal mode” with their analytical approach and sourcing. Your consumer AI handles the interface and personalization, but it channels different news organizations’ editorial identities.

News organizations become editorial consultants to your personal AI. Their value-add becomes their perspective and credibility, not just raw information. You might even ask explicitly: “Give me the Reuters take on this story” or “How would the Financial Times frame this differently?”

The Real Prize: Cognitive Monopoly

But news is just one piece of a much larger transformation. Your Knowledge Navigator won’t just fetch information—it will manage your calendar, draft your emails, handle your shopping, mediate your social interactions, filter your dating prospects, maybe even influence your political views.

Every interaction teaches it more about you. Every decision it helps you make deepens the relationship. The switching costs become enormous—it would be like switching brains.

This is why the current AI race looks almost quaint in retrospect. We’re not just competing over better chatbots. We’re competing to become humanity’s primary cognitive interface with reality itself.

The Persona Moat

Remember Theo in Her, falling in love with his AI operating system Samantha? Once he was hooked on her personality, her way of understanding him, her unique perspective on the world, could you imagine him switching to a competitor? “Sorry Samantha, I’m upgrading to a new AI girlfriend” is an almost absurd concept.

That’s the moat we’re talking about. Not technical superiority or feature sets, but intimate familiarity. Your Knowledge Navigator will know how you think, how you communicate, what makes you laugh, what stresses you out, how you like information presented. It will develop quirks and inside jokes with you. It will become, in many ways, an extension of your own mind.

The economic implications are staggering. We’re not talking about subscription fees or advertising revenue—we’re talking about becoming the mediator of trillions of dollars in human decision-making. Every purchase, every career move, every relationship decision potentially filtered through your AI.

Winner Take All?

The switching costs suggest this might be a winner-take-all market, or at least winner-take-most. Maybe room for 2-3 dominant Knowledge Navigator platforms, each with their own personality and approach. Apple’s might be sleek and privacy-focused. Google’s might be comprehensive and data-driven. OpenAI’s might be conversational and creative.

But the real competition isn’t about who has the best underlying models—it’s about who can create the most compelling, trustworthy, and irreplaceable digital relationship.

What This Means

If this vision is even partially correct, we’re watching the birth of the most valuable companies in human history. Not because they’ll have the smartest AI, but because they’ll have the most intimate relationship with billions of people’s daily decision-making.

The Knowledge Navigator wars haven’t really started yet. We’re still in the pre-game, building the underlying technology. But once personal AI becomes truly personal—once it knows you better than you know yourself—the real competition begins.

And the stakes couldn’t be higher.

The Algorithmic Embrace: Will ‘Pleasure Bots’ Lead to the End of Human Connection?

For weeks now, I’ve been wrestling with a disquieting thought, a logical progression from a seemingly simple premise: the creation of sophisticated “pleasure bots” capable of offering not just physical gratification, but the illusion of genuine companionship, agency, and even consent.

What started as a philosophical exploration has morphed into a chillingly plausible societal trajectory, one that could see the very fabric of human connection unraveling.

The initial question was innocent enough: In the pursuit of ethical AI for intimate interactions, could we inadvertently stumble upon consciousness? The answer, as we explored, is a resounding and paradoxical yes. To truly program consent, we might have to create something with a “self” to protect, desires to express, and a genuine understanding of its own boundaries. This isn’t a product; it’s a potential person, raising a whole new set of ethical nightmares.

But the conversation took a sharp turn when we considered a different approach: treating the bot explicitly as an NPC in the “game” of intimacy. Not through augmented reality overlays, but as a fundamental shift in the user’s mindset. Imagine interacting with a flawlessly responsive, perpetually available “partner” whose reactions are predictable, whose “needs” are easily met, and with whom conflict is merely a matter of finding the right conversational “exploit.”

The allure is obvious. No more navigating the messy complexities of human emotions, the unpredictable swings of mood, the need for compromise and difficult conversations. Instead, a relationship tailored to your exact desires, on demand, with guaranteed positive reinforcement.

This isn’t about training for better human relationships; it’s about training yourself for a fundamentally different kind of interaction. One based on optimization, not empathy. On achieving a desired outcome, not sharing an authentic experience.

And this, we realized, is where the true danger lies.

The ease and predictability of the “algorithmic embrace” could be profoundly addictive. Why invest the time and emotional energy in a flawed, unpredictable human relationship when a perfect, bespoke one is always available? This isn’t just a matter of personal preference; on a societal scale, it could lead to a catastrophic decline in birth rates. Why create new, messy humans when you have a perfectly compliant, eternally youthful companion at your beck and call?

This isn’t science fiction; the groundwork is already being laid. We are a society grappling with increasing loneliness and a growing reliance on digital interactions. The introduction of hyper-realistic, emotionally intelligent pleasure bots could be the tipping point, the ultimate escape into a world of simulated connection.

The question then becomes: Is this an inevitable slide into demographic decline and social isolation? Or is there a way to steer this technology? Could governments or developers introduce safeguards, programming the bots to encourage real-world interaction and foster genuine empathy? Could this technology even be repurposed, becoming a tool to guide users back to human connection?

The answers are uncertain, but the conversation is crucial. We stand at a precipice. The allure of perfect, programmable companionship is strong, but we must consider the cost. What happens to society when the “game” of connection becomes more appealing than the real thing? What happens to humanity when we choose the algorithmic embrace over the messy, complicated, but ultimately vital experience of being truly, vulnerably connected to one another?

The future of human connection may very well depend on the choices we make today about the kind of intimacy we choose to create. Let’s hope we choose wisely.