ON_Discourse is a membership based community of C-suite leaders, investors, and innovators focused on driving business transformation through the discipline of discourse.
Learn More
NEWSLETTER
The next leap in AI will come from design.
The problem with AI isn’t the ambition as much as the language we use to describe it. It’s all fog, no form. And when the language is that fuzzy, design has nothing to grab onto.
PODCAST
Episode #021
Toby and Chmiel continue last week’s conversation on human ingenuity with a founder who’s putting it to the test. Katherine von Jan (KVJ to those who know her) joins the pod to talk about her new venture, Tough Day, and its mission to unlock the best of human potential in an AI-saturated workplace. As a longtime ON_Discourse member and former Salesforce exec turned startup founder, KVJ has a sharp perspective on where AI helps, where it hurts, and why the future of work depends on not automating away what makes us human.
LISTEN
NEWSLETTER
Ingenuity is the last unfair advantage.
AI isn’t a secret weapon anymore. Everyone has it. The edge isn’t in how many prompts you can fire off or how many models you can chain together. The real advantage comes from what you, uniquely, are willing to see, risk, and create when the tool runs out of patterns to copy.
PODCAST
Episode #020
Toby, Matt, and Dan wrestle with what it means to have a “voice” in an AI-mediated world, and what happens when that voice gets flattened, filtered, or faked. This episode digs into the paradox of human ingenuity in the age of synthetic output, the strange intimacy of talking to machines, and why friction, not fluency, might be the future of AI design.
LISTEN
NEWSLETTER
AI’s job isn’t to be right. It’s to make sure you know what you believe, and why.
We need to start designing for useful friction, not automating it away. That doesn’t mean building smarter models, it means building systems that help humans see where their reasoning holds, where it wobbles, and where it’s been unconsciously shaped by someone else’s priors.
PODCAST
Episode #019
Toby, Dan, and guest Craig Hepburn step inside the AI browser wars, and ask if the future of the internet is being quietly rewritten not by models, but by interfaces. This episode zooms in on Perplexity’s new agentic browser Comet, the real promise (and limits) of context-aware assistants, and what gets lost when AI "helps" too much.
LISTEN
NEWSLETTER
Discomfort is the unlock to your best work, if you have the courage to lean into it.
Discomfort is the source of human ingenuity. The bridge between taste and execution. The thing that makes good work better, and better work unforgettable.
We’ve been offloading cognitive tasks for centuries, from cave paintings to calendars to Google Maps. But as AI steps into more sophisticated roles in writing, decision-making, and even ideation, we’re entering new territory. Are we slowly anesthetizing the very muscle that makes us human?
Matt Chmiel
PODCAST
Episode #018
Toby, Dan, and Chmiel interrogate the quiet crisis hiding inside the AI boom: we’re offloading too much thinking and calling it strategy. This episode explores what gets lost when cognitive offloading becomes the norm, and why the future of leadership may depend on friction, not flow.
LISTEN
NEWSLETTER
AI can help you think faster. But only humans will tell you when you're wrong.
Leadership class doesn’t need more tools. It needs a network of trusted humans who can challenge assumptions, stress-test ideas, and stay in the room long enough to make the work better. Not just right, but real.
The Promise, the Threat, and the Mirror
ON_Discourse Editorial Team
06•26•25
NEWSLETTER
If AI can generate your content, can it generate your taste?
If speed becomes the creative edge, what happens to deliberation, serendipity, or even procrastination? If your AI co-pilot can simulate every outcome, do you still have the guts to follow your gut? And if AI can remix a million influences at once, is taste still a human skill, or just another variable in the model?
A Robber Baron’s Guide to AI-Era Media
Eric Gillin
06•13•25
This won’t work
I can think of at least 6 reasons why none of this could ever happen.
Tony Haile
06•17•25
NEWSLETTER
What is your personal moat?
In a world where anything can be cloned, shipped, or scaled in seconds, your moat isn’t what you build, it’s how you adapt. Let me turn that platitude into something even more cringe: Soul becomes strategy. Curiosity becomes infrastructure. What is your moat?
PODCAST
Episode #017
Dan, Toby, and Chmiel go behind the scenes of Cannes Lions to unpack the real programming, posturing, and potential shaping the creative industry’s biggest week. Is the future of advertising being written on the main stage or whispered behind closed doors?
LISTEN
NEWSLETTER
What if the idea you cringed at was actually the beginning of something true?
The AI transformation everyone is talking about - including us, including me every week - is not here yet. Therefore, this week’s newsletter edition acknowledges the impatience we are all feeling about AI, and it also credits the crazy thinkers who are pushing us into uncomfortable cognitive places.
PODCAST
Episode #016
We’re joined by Taylor Halsted of Shamrock Capital to interrogate the illusion of AI efficiency. What if the mess we’re trying to automate out of work is actually where creativity, connection, and culture live?
LISTEN
NEWSLETTER
Is MCP kicking you off the internet?
MCP stands for Model Context Protocol. And if it lives up to the ambition behind it, it will become one of the most disruptive technologies in the history of the internet. It wants to turn AI into an interface. One that replaces search. Replaces apps. And ultimately replaces you (specifically your demands, behaviors, actions, clicks, searches, and swipes).
PODCAST
Episode #015
Toby sits down with Henrik Werdelin and Nicholas Thorne, co-authors of Me, My Customer, and AI, to interrogate the optimistic thesis at the heart of their book, and the paradoxes it reveals. From donkeycorns to Dunbar numbers, this session explores what happens when everyone is an entrepreneur, and no one stands out.
LISTEN
PODCAST
Episode #014
Toby and Dan track the decline of Google Search, the end of SEO, and what’s replacing it: group chats, agentic systems, and personalized internets built around behavior, not clicks.
LISTEN
NEWSLETTER
Can fake empathy hurt us for real?
LLM agents have been trained to perform a very impressive parlor trick. They can empathetically convert the context, subtext, pretext, and plain text of all of our individual human prompts into shockingly relevant experiences. It feels like a miracle but it’s really just code.
NEWSLETTER
SEO still works, but does it still matter?
Little by little, prompting is replacing searching. The cracks are starting to creep into the discourse. Executives in media and marketing are starting to see the old rules lose their sway.
PODCAST
Episode #013
We have been hosting closed-door, private Group Chats at ON_Discourse for 2 years. In this episode, Toby and Chmiel talk about the ON_Discourse version of group chats and how they are different than the contentious, political experience that was recently profiled by Ben Smith at Semafor.
LISTEN
How Mark Zuckerberg Became a Shape-Shifter Across Culture—and What That Tells Us About Power Now
Toby Daniels
05•06•25
NEWSLETTER
What if Jack Dorsey is right?
Dorsey’s provocation invites an exploration on the future of IP law. If we don't rethink the purpose of intellectual property, we aren't protecting creativity anymore. We’re protecting a future that no longer exists.
Gen Z isn’t coming for the internet. They’re rebuilding it from scratch. And most of us haven’t noticed.
Toby Daniels
04•25•25
NEWSLETTER
What is the rush?
The most interesting builders in our network aren’t just moving fast. They’re using speed to reshape intention. To prototype purpose. To rediscover why they’re building in the first place
PODCAST
Episode #012
Toby and Dan turn a weekly meeting into an impromptu podcast recording about the future of digital strategy and why notifications suck.
LISTEN
Gen Z isn’t coming for the internet. They’re rebuilding it from scratch. And most of us haven’t noticed.
Toby Daniels
04•21•25
NEWSLETTER
AI isn’t revolutionary to Gen Z.
We invited two teenagers into a group chat. No titles. No decks. No “vision.” Just two high schoolers who are already using AI tools the way we once used FTP clients and WordPress installs. They’re building with no reverence whatsoever.
PODCAST
Episode #011
Toby interviews a couple of teenagers who are vibecoding their way through the modern internet. This conversation reveals an alternative perspective on the AI revolution, from a generation that never knew a world without the web.
LISTEN
NEWSLETTER
AI will commoditize democratization
In the last two years, we’ve hosted hundreds of events about AI and nearly every one is centered on the idea that AI will democratize creativity, coding, knowledge, etc. But is it true? And what does it really mean?
NEWSLETTER
The Hardest Part Isn’t Building AI — It’s Selling It
Our members are at the cutting edge of AI. Their prospective clients see the promise of AI - they see the same headlines as everyone else - but they don’t yet know how to use it, where it fits, or what it’s worth. The friction is not resistance. It’s uncertainty.
For decades it has been the go-to cliché in tech. Not only does this lead to boring, useless perspectives, but it also relies on an outdated research methodology that makes no sense.
Toby Daniels
03•17•25
NEWSLETTER
It’s Time to Build
Many of our members tell us that this moment feels like the 90’s all over again. Anything is possible because anyone can make anything. All you need is a voice and some vibes.
POWERED BY DECODER • EDITED BY CHMIEL
Discourse in the News
Our top agent identified three provocations for the week of March 3, 2025
1. Consumer Agents are Coming!
Amazon and T-Mobile (Telekom) made two separate announcements about AI-powered personal assistants in the upgraded Alexa+ and the new so-called AI-Phone.
These agents won’t just respond to consumer prompts; they will autonomously act and manage daily tasks without a command.
READ THESE FIRST
Telekom, Perplexity debut “AI Phone” The Rundown AI, Rowan Cheung
Alexa+ tracks your WHOLE life Superhuman AI, Zain & team
YOU TELL US
Are agentic devices worth the privacy compromise?
Is this just marketing?
2. Is GPT4.5 a Big Emotional Whiff?
GPT4.5 was released with improved emotional intelligence, astronomical pricing and marginal improvements. Many reviewers were unimpressed.
READ THESE FIRST
OpenAI’s GPT4.5 is here The Rundown AI, Rowan Cheung
GPT4.5 is confusing Ben’s Bites Newsletter
YOU TELL US
Is better emotional intelligence worth 30x higher usage fees?
Are new model releases boring now?
3. Is Inference Time Deeper Thinking?
New “deep research” models spend extra “thinking time” to break down and process prompts that rival human thought. What does this updated UI really mean?
READ THIS FIRST
Think Before Your Next Move: A Celebration of Inference Time Compute and AI Nate’s Substack
YOU TELL US
Does watching a model break down and “think” through a prompt enhance your research?
Is this a gimmick?
