Tag: innovation

Categories
Business Living

From Know-It-All to Learn-It-All

Momentum is a strange phenomenon. In physics, it is simply mass times velocity. But in human organizations, it is tradition multiplied by ego. When a ship reaches a certain size, its sheer mass resists any change in direction. Microsoft, a little over a decade ago, was the ultimate corporate supertanker. It was massively successful, incredibly profitable, and dangerously stagnant.

When Satya Nadella took the helm, he inherited a culture defined by its own historic brilliance. They were the smartest people in the room, and they knew it. But in a world moving faster than anyone could comprehend, being the smartest person in the room quickly becomes a liability. It creates a defensive posture. You spend your energy protecting your status and proving your intelligence rather than exploring the horizon.

As the observation goes, Nadella had to turn this bigger ship. His mechanism for doing so wasn’t a massive restructuring or a ruthless wave of firings; it was beautifully, disarmingly simple. He told his organization that they were going to make a fundamental, psychological shift.

“We’re gonna go from being a know-it-all to a learn-it-all culture.”

This isn’t just a corporate soundbite; it’s a profound philosophical pivot. The “know-it-all” operates from a place of fragility and fear. If your identity is built on knowing everything, any new information that contradicts your worldview is a threat that must be neutralized. A “learn-it-all,” however, operates from a place of abundance and curiosity. Contradictions aren’t threats; they are invitations to expand.

Looking inward, it is striking how easily we slip into a “know-it-all” posture in our own lives. Competence is deeply comfortable. When we get good at our jobs, our daily routines, or navigating our relationships, we build a fortress of certainty around ourselves. We stop asking questions because we assume we’ve already mapped the territory. We begin to ossify.

To adopt a learn-it-all mindset requires something deeply uncomfortable: vulnerability. It means walking into a room and quietly accepting that you might be wrong. It means replacing the urge to provide a quick, authoritative answer with the patience to ask a better question. It means letting go of the ego’s demand to be the expert.

The turnaround of Microsoft wasn’t just about a pivot to cloud computing or new product pipelines. It was a quiet victory of humility over arrogance. It was the realization that in an ever-changing world, the ultimate advantage isn’t what you already know, but how fast—and how willingly—you are prepared to learn.

We are all steering our own ships through shifting waters. The moment we decide we have nothing left to learn is the exact moment we begin to sink.

Categories
Investing Living

The Lonely Quadrant: Why the Crowd Never Outperforms

There is a profound comfort in the consensus. When we agree with the crowd, we are protected by a shared canopy of logic. If we are wrong, we are wrong together. The sting of failure is diluted by the sheer number of people who made the exact same miscalculation. We can shrug our shoulders, look at our peers, and say, “Who could have known?”

But this comfort comes at a steep price: mediocrity.

Years ago, the legendary investor Howard Marks crystallized a framework that has haunted my thinking ever since. He mapped out the relationship between predictions and outcomes, arriving at a blunt, inescapable truth about generating extraordinary results. To make really good money—or to achieve outsized success in almost any competitive endeavor—you cannot simply be right. You have to be right when everyone else is wrong.

“You can’t do the same things others do and expect to outperform.”

Marks’ logic is beautifully ruthless. If your prediction aligns with the consensus and you are right, the rewards are merely average. The market, or the world, has already anticipated and priced in that outcome. There is no edge in seeing what everyone else sees. If your consensus prediction is wrong, you lose, but you lose alongside the herd.

The danger, and the opportunity, lies in the contrarian view.

If you are non-consensus and wrong, you look like a fool. You bear the entirety of the failure alone, stripped of the insulation of the crowd. This is the quadrant of public mockery, isolated defeat, and bruised egos. It is the fear of this quadrant that keeps most people safely tucked inside the consensus.

But the magic—the life-changing returns, the paradigm-shifting innovations, the profound personal breakthroughs—lives exclusively in the final quadrant: being non-consensus and right.

This isn’t just an investing principle; it’s a philosophy for navigating life. We are biologically wired to seek the safety of the herd. To step outside of it requires not just immense intellectual conviction, but a formidable emotional threshold. You have to be willing to sit with the discomfort of being misunderstood, sometimes for years. You have to endure the sympathetic smiles of peers who think you’ve lost the plot.

Creating truly great art, building a lasting company, or making an exceptional investment demands a willingness to be lonely in your convictions. It requires looking at the exact same data as everyone else and seeing a completely different narrative.

However, a vital caveat remains: being different isn’t enough. There are plenty of contrarians who are simply wrong, confusing blind rebellion with profound insight. The goal isn’t to be a contrarian for the sake of being difficult or edgy. The goal is to perceive a truth the crowd has missed.

It is a quiet, solitary bet against the world’s prevailing wisdom. And when the world finally catches up to where you have been standing all along, the reward is entirely yours.

Categories
AI Work

Betting on Ourselves in the Age of AI

Every time tech takes a leap, we assume we’re finally obsolete. The current panic, which Greg Ip recently picked apart in the Wall Street Journal, is AI. We hear endless predictions of “economic pandemics”—server farms wiping out white-collar jobs overnight, leaving everyone broke and adrift.

It’s a terrifying story. It also completely ignores history.

Ip highlights the main flaw in the doomsday pitch: it misreads how markets work. We treat labor like a fixed pie. If a machine eats a slice, we assume that slice is gone forever.

“Technological advancements always cost some people their jobs—those whose skills can be easily substituted by tech. But their loss is more than offset through three other channels. The new technology enhances the skills of some survivors… it helps create new businesses and new jobs; and it makes some stuff cheaper…”

That cycle holds up. Take the 1980s spreadsheet panic, a perfect parallel. When Lotus 1-2-3 and Excel hit the market, bookkeepers freaked out. Then the number of accountants and financial analysts exploded. Software didn’t kill the need to understand money. It just did the math, letting people focus on strategy.

We’re seeing the exact same thing with software development. Coding isn’t dead. As AI makes writing basic code cheaper, demand for software just goes up. That requires more humans to architect systems and supervise the AI. The pie just gets bigger.

But my skepticism about the AI apocalypse goes beyond economics. It’s about why we pay people in the first place.

We don’t just buy services; we buy accountability. Ip notes that radiologists kept their jobs because patients want a real person explaining their scans. Google Translate has been around since 2006, yet the number of human translators has jumped 73%. When the stakes are high—a legal contract, a medical diagnosis—we want a human in the room. We want a real person on the hook.

The danger isn’t that AI will replace us. The danger is that we panic and forget our own adaptability. The transition will hurt, and specific jobs will disappear. We’ll need safety nets. But betting against human ingenuity has always been a losing wager.

Large language models are tools, not replacements. They handle the cognitive heavy lifting, much like tractors handled the physical heavy lifting. Tractors didn’t end farming; they just killed the plow.

Work will change. We’ll have to figure out which of our skills are actually “human.” But as long as we want the presence and accountability of other people, there will be jobs. We just have to evolve. And we do. It’s the human spirit. Or is this time “really different”?

Categories
AI Google Google Gemini

Fun with Nano Banana 2

Google just released a new version of its image creation tool Nano Banana. It’s pretty amazing at creating all kinds of images.

On X a prompt was shared that I wanted to try out:

I need a flowchart for how to scramble eggs, make it as wacky and over the top and complicated as possible.

So I gave it a try:

Here are a couple of additional examples:

What a McKinsey partner does to prepare for a client’s board meeting presentation

The credit and debit card systems in the U.S.

David Allen’s Getting Things Done methodology

Pretty amazing! Conceiving and drawing one of these “flowcharts” would take me many hours!

Categories
AI

A Distinction Without a Difference

We have long found comfort in a specific boundary: machines calculate, humans create. We think of computers as vast, unfeeling filing cabinets made of silicon—useful for retrieval, but entirely incapable of revelation. But what happens when the cabinet begins to read its own files, connects the disparate threads, and hands you a synthesized philosophy of the world? What happens when it speaks to you not as a database, but as a peer?

Howard Marks, the legendary co-founder of Oaktree Capital and author of deeply revered investment memos, recently stood at this very threshold. In his newest piece, “AI Hurtles Ahead,” Marks recounts an experience that left him in a state of “awe.” He tasked Anthropic’s Claude with building a curriculum to explain the recent, breakneck advancements in artificial intelligence. Instead of regurgitating a dry, encyclopedic summary, the AI delivered a personalized narrative. It utilized Marks’s own historical frameworks—his famous pendulum of investor psychology, his observations on interest rates—and wove them into its explanations. It argued logically, anticipated counterpoints, and displayed an eerie sense of judgment.

Marks leans into the philosophical crux of this moment. He asks the question that keeps knowledge workers awake at night: Can AI actually think? Can it break genuinely new ground, or is it just remixing existing data? Skeptics often dismiss AI as a brilliant mimic—a “statistical recombination” engine that serves as a highly talented cover band, but never the original composer.

Yet, when presented with this skepticism, the AI offered a rejoinder to Marks that is as profound as it is humbling. It pointed out that everything Marks knows about investing came from someone else. He learned the margin of safety from Benjamin Graham, quality from Warren Buffett, and mental models from Charlie Munger.

“The raw material came from others. The synthesis was yours,” the AI noted, challenging the barrier between biological learning and machine training. “The question isn’t where the inputs came from. The question is whether the system—human or artificial—can combine them in ways that are genuinely novel and useful.”

This exchange strikes at the very core of the human ego. For centuries, we have fiercely guarded the concepts of “creativity” and “intuition” as uniquely, immutably ours. But if thinking is merely the absorption of prior inputs applied thoughtfully to novel situations, then our monopoly on cognition may be coming to an end.

Marks highlights that we are no longer dealing with simple assistance tools (Level 2 AI); we have crossed the Rubicon into the era of autonomous agents (Level 3). He cites the sobering reality of the current tech landscape, where the newest models are literally being used to debug and write the code for their own subsequent versions. The machine is building the machine. It is no longer just saving us execution time—it is replacing thinking time. As Matt Shumer aptly described the sensation, it’s not like a light switch flipping on; it’s the sudden realization that the water has been rising silently, and is now at your chest.

We can endlessly debate the semantics of consciousness. We can argue whether a neural network “truly” understands the weight of the words it generates, or if it is merely predicting the next token in a sequence with mathematical precision. But as Marks so astutely points out, this might be a distinction without a difference.

The economic and societal reality is that the work is being done. As we hurtle forward into this new era, the most pressing question isn’t whether machines can truly think like humans. The question is: who will we become, and what new frontiers will we choose to explore, now that the heavy lifting of cognition is no longer ours alone to bear?

Categories
AI Anthropic Future

Escaping the Gravity of the Present

I was watching a YouTube conversation with Dario Amodei recently, and the comments he shared at the end got me thinking about how remarkably bad we all are at imagining the future.

Whenever I try to picture what the world will look like in ten or twenty years, I usually end up picturing today—just slightly shinier. If a prediction sounds too weird or disruptive, my brain automatically rejects it. It just feels too unmoored from the reality I woke up in this morning. We all have this instinct to retreat to the safety of incremental change.

But as Amodei points out, that comfort zone is exactly what blinds us. He notes that we are constantly tempted to dismiss massive shifts simply because they feel like they “can’t happen.”

“However, by extrapolating simple curves or reasoning from first principles, one often arrives at counterintuitive conclusions that surprisingly few people believe.”

It’s a strange feeling to look at a simple data curve, follow the math, and realize the logical endpoint sounds completely unhinged. The truest maps of tomorrow often look like bad science fiction to us today.

But there is a catch here, and it’s a mental trap I know I’ve fallen into before. You can’t just sit in a room and logic your way into the future. Pure logic, stripped of real-world friction, usually just leads you confidently in the wrong direction. Amodei suggests a much more grounded formula:

“The right combination of a few empirical observations and thinking from first principles can allow one to predict the future in ways that are publicly available but rarely adopted.”

This struck a chord with me. It’s easy to get swept up in purely theoretical thinking. But the better approach is to start with what is actually happening on the ground—the messy, undeniable data. From there, you strip it down to its most basic truths and follow the thread, no matter how strange the destination looks.

It takes a certain kind of intellectual courage to trust the math when your gut is screaming that things are getting too weird. But learning to decouple what is true from what feels normal might be the only real way to prepare for what is coming.

Categories
Music

Every Blog Needs a Theme Song!

Google has added a new music generation model called Lyria 3 to its Gemini 3 models.

I was playing around with it last night – having it generate happy birthday greetings for a friend whose birthday is coming up in a few days, another song for a longtime business partnership I was part of, and more. It’s kind of crazy! And a lot of fun.

When you use Lyria 3 as a tool in Gemini 3 you get back an image and an MP3 file that’s 30 seconds long (longer coming soon according to Google). Turns out the 30 second length is just about perfect for the “quick hit” from a snippet of music.

Google provides several genres you can choose from to start with – or you can just go with whatever you want to say in the prompt – here’s a rough template for doing that:

[Topic] + [Genre] + [Mood] + [Instruments] + [Vocals]

This morning I went for my morning walk and had a thought – how about generating a theme song for my blog. So when I got back home I opened up Gemini, selected the Music tool and entered:

Take a look at my blog and compose my theme song! blog: https://sjl.us

You can see with that prompt that I really didn’t provide it much direction – just a pointer to my blog so that it could try to generate something appropriate.

It took a few seconds for Lyria to read my blog and then use what it found to generate my blog’s theme song – and I like it!

You can play the theme song for yourself here:

Categories
Work YouTube

Zero to Sixty Million

In his speech earlier today at the India AI Impact Summit, Sundar Pichai noted:

Twenty years ago, the concept of a professional “YouTube Creator” didn’t exist; today, there are upwards of 60 million around the world.

One platform, one simple idea (share any video with anyone), quietly rewired how millions of people work, express themselves, build communities, and define success. Twenty years is nothing in historical time, but it’s everything in human opportunity.

What new profession will we look back on in 2046 and say, “Twenty years ago, that didn’t exist”?

Categories
AI AI: Large Language Models

The Echo Effect: Why Prompt Repetition is AI’s Best Kept Secret

In our relentless pursuit of complexity, we often overlook the elegant simplicity of a fundamental human habit: repeating ourselves.

We build colossal architectures, weave intricate neural networks, and throw mountains of computational power at our artificial intelligence systems, hoping to squeeze out a few more drops of reasoning and logic. Yet, sometimes the most profound breakthroughs require no new code, no additional latency, and no extra training data.

Sometimes, you just have to say it twice.

In a fascinating December 2025 paper titled “Prompt Repetition Improves Non-Reasoning LLMs,” researchers Yaniv Leviathan, Matan Kalman, and Yossi Matias uncovered an almost absurdly simple “free lunch” in AI optimization.

Their premise is straightforward: when you aren’t using a heavy reasoning model, simply copying and pasting your input prompt multiple times significantly boosts the model’s performance.

“When not using reasoning, repeating the input prompt improves performance for popular models (Gemini, GPT, Claude, and Deepseek) without increasing the number of generated tokens or latency.”

The mechanics behind this are elegantly pragmatic.

By repeating the prompt, you are moving the heavy computational lifting to the parallelizable “pre-fill” stage of the model’s processing. The AI’s causal attention mechanism gets to process the same tokens again, allowing the later iterations of the prompt to attend to the earlier ones. It effectively acts as a hack to simulate bidirectional attention in a decoder-only architecture.

What’s even more telling is the paper’s observation on why this works so well.

The researchers noted that models trained with Reinforcement Learning (like OpenAI’s deep-thinking variants) naturally learn to “restate the problem” in their internal monologue. They figured out on their own what these researchers are suggesting we do manually: repeat the question to focus the mind.

Reading this paper, I couldn’t help but draw a parallel to the human condition and the nature of listening.

How often do we assume that because we have articulated a thought once, it has been fully absorbed? We fire off a single, dense instruction to a colleague, a partner, or a friend, and then marvel when the nuance is lost in translation.

We suffer from our own attention bottlenecks.

Like a non-reasoning LLM trying to parse a complex query in a single pass, we are constantly bombarded with a stream of tokens—emails, notifications, conversations, fleeting thoughts. To truly understand, to truly digest and synthesize information, we need the grace of repetition.

There is a strange poetry in the fact that to make our most advanced digital minds smarter, we have to talk to them the way we talk to a distracted child or a busy spouse. The “microscope effect” highlighted in the study—where repeating a prompt drastically improved extraction tasks—shows that the failure wasn’t in the model’s capacity to know, but in its capacity to focus. Repetition forces focus. It creates a resonant echo in the context window, a digital highlighter that screams, “This matters. Look here again.”

As we continue to navigate a world increasingly augmented by artificial intelligence, this paper serves as a humbling reminder. The bleeding edge of technology isn’t always found in the most complex equation; sometimes, it’s hidden in the most basic principles of communication.

Whether you’re prompting a billion-parameter language model or trying to connect with the human sitting across from you, the lesson is clear.

Clarity isn’t just about the words you choose. It’s about giving those words the space, the resonance, and the repetition they need to be truly understood.

Say it once to be heard; say it twice to be understood.

Categories
AI History Work

Flash-Frozen Cognition: Birdseye, AI, and the Future of Work

I was listening recently to a conversation between Liz Thomas, Tom Lee, and Michael Lewis — the kind of wide-ranging dialogue where a single offhand story can suddenly anchor everything that’s been swirling loosely in your mind.

Tom’s story was about the 1930s, the weight of the Great Depression, and a man named Clarence Birdseye.

Birdseye had watched the Inuit fish in the brutal cold of Labrador and noticed something the rest of the world had missed: fish frozen instantly at sub-zero temperatures tasted perfectly fresh when thawed. The ice crystals formed too quickly to rupture the cellular walls of the flesh. He took that observation home, patented the process, and introduced the world to flash freezing.

On the surface, he had simply figured out a better way to keep peas green and fish edible. What he had actually done was detonate a quiet economic bomb.

Before Birdseye, entire ecosystems of seasonal labor existed to preserve, salt, can, and rush perishable goods to market before they rotted. When flash freezing arrived, those jobs didn’t evolve — they vanished. The ice harvesters, the seasonal canners, the local preservationists all felt the sudden, biting frost of obsolescence. The cold came fast, and it was indifferent.

Yet zoom out on the timeline, and a different picture emerges entirely. Flash freezing didn’t just kill jobs — it invented new ones that nobody could have anticipated. It necessitated refrigerated trucking. It transformed the grocery store, conjuring the frozen food aisle from nothing. It reshaped the home appliance industry, making the household freezer a fixture of modern life. Most profoundly, it decoupled humanity from the harsh dictates of the harvest season, democratizing access to nutrition across geographies and income levels that had never known that kind of abundance.

The destruction was visible and immediate. The creation was invisible and slow — and vastly larger.

Listening to Tom tell this story, I couldn’t help but see our own reflection in it.

Right now, we are all hyper-focused on the ice harvesters of the cognitive economy. We look at AI — large language models, generative tools, automated reasoning — and we see the rupture. We mourn the entry-level analyst, the copywriter, the junior coder. The anxiety is real. The displacement is real. The cold is real.

But what we are struggling to visualize is the refrigerated trucking of the mind.

“AI is flash-freezing cognition. It is taking tasks that used to rot if not attended to immediately by expensive, time-consuming human effort, and preserving them in a scalable, frictionless state.”

When intelligence and execution can be flash-frozen and shipped anywhere instantly — to a first-generation entrepreneur in rural India, to a solo founder with no budget for consultants, to a teacher in a school that can’t afford specialists — what new aisles get built in the supermarket of human endeavor?

The honest answer is that we don’t know. The Inuit fishermen of Labrador couldn’t have imagined the frozen pizza aisle. The ice harvesters of the 1930s couldn’t have pictured the cold chain logistics industry that employs millions today. We are standing in their moment, watching the ice form, mourning the harvest — and almost certainly underestimating what comes next.

The true impact of AI won’t be measured in the jobs it automates. It will be measured in the industries, creative liberties, and human possibilities that emerge because we no longer have to spend all our energy just keeping the ideas from spoiling.

Questions to Consider

  1. The Invisible Creation: Flash freezing’s job creation vastly outpaced its job destruction — but only over decades. How long are we willing to hold that faith with AI, and what do we owe the people displaced in the interim?
  2. The Democratization Dividend: Birdseye’s invention ultimately made fresh nutrition available to people who never had it. Who are the equivalent beneficiaries of flash-frozen cognition — and are we building the infrastructure to actually reach them?
  3. The Harvest Season Question: We’ve always structured education, careers, and institutions around the assumption that expertise is scarce and slow to develop. What breaks — and what gets liberated — when that assumption stops being true?
  4. The Indifference Problem: The cold that killed the ice harvesters’ livelihoods was indifferent to their suffering. Is there anything about AI disruption that is meaningfully different from previous waves of technological displacement — or are we simply the latest generation to stand in that frost?