Robert Solow said something in 1987 that feels written for 2026: "You can see the computer age everywhere but in the productivity statistics." We are living the AI version of it right now.
When AI becomes Your Therapist
In 2013, Her showed us a man falling in love with his AI operating system. We watched it as science fiction—a little unsettling, but safely distant. That was 13 years ago. Today, people are using AI for therapy. For companionship. For the kind of emotional support we used to get only from other humans.
The Great Inversion: When Articulation Becomes the Bottleneck
25 years ago, Linus Torvalds said: "Talk is cheap. Show me the code." In 2026, Zerodha's CTO flipped it: "Code is cheap. Show me the talk." The great flip happened in just 25 years.
2025, or What Happened When the Reservoir Ran Dry
When I started this newsletter (back in 2024), I had the confidence of someone who'd never run out of ideas. By this year, the reservoir had run dry. But I kept writing anyway. Here's what I learned from writing 51 newsletters in 2025: → AI dominated my output (31%) when I planned to write about policy → Being right about content marketing's comeback didn't prevent months of struggle → The newsletter became less about expertise, more about documenting uncertainty → Writing became thinking in public The biggest lesson? Markets can remain irrational longer than you can remain solvent. And that applies to career too.
The Great Inversion: How AI Flipped Copyright Inside Out
In 2020, my manager rejected a candidate for plagiarism. In 2024, Anthropic paid $1.5 billion for using pirated books to train Claude, even though a judge ruled the actual training was "fair use."
Why CORE Traits Matter More Than Skills in the AI era
We posit it's because of CORE traits: 4 unchanging human qualities that will outlast every technical skill you'll ever learn. It's not nepotism, favoritism, or flattery. It's CORE traits that don't show up on resumes. And in the AI era, these traits matter more than ever
Why Retrieval Practice Beats Endless Study
The ability to pull information from memory in a challenging environment > Constant reading, reviewing and preparing tonnes of notes.
The End of Pretend Work: When AI Exposes What Was Never There
For decades, white-collar work consisted of generating impressive-looking documents that few people read carefully and even fewer verified. This work felt productive. It kept people busy. It generated revenue. It filled time. This is why I call AI a white-collar revolution. It’s doing what the industrial revolution did to many blue collar jobs of its time. It changed what constituted value and forced people to add value in new ways.
Why Arattai’s Positioning Matters And India’s Product Ambition Gap
I analyze why Koo and Hike failed despite millions of users. And why Zoho's Arattai may be repeating their mistakes. I argue that, at its core, it’s the difference between "Made FOR India" and "Made FROM India." To be clear: Zoho is uniquely positioned to break this pattern. They've proven they can compete globally in B2B. The question is whether Arattai's positioning reflects that ambition.
The Writing Looks ChatGPTish
When people say they can spot AI writing from a mile, what do they actually mean? Excessive em dashes? Short paragraphs? Words like "delve" and "meticulous"? LinkedIn was full of cringe long before ChatGPT. AI didn't invent buzzwords or performative prose. The algorithm simply picked what went viral. We're pretending that before ChatGPT, all writing flowed from pure creativity. As if "On Writing Well" and "The Elements of Style" never existed. We always followed rules. But our inability to follow them perfectly made our writing unique. AI follows rules perfectly. That's the problem. I explore this paradox in this essay.
Musings by Madhav 