Everyone says AI is eating SaaS. But when I look at the people and companies around me, I don’t see many cancelled subscriptions. Sure there are a few on the margins, but what’s really going on here?

The public numbers provide a strong counter-narrative. SEC filings for 14 of the largest SaaS companies (ex: Salesforce, Adobe, ServiceNow, Shopify, etc) from Q1 2024 through mid 2026 show not one reporting year-over-year revenue declines in any quarter. Stocks have fallen based on investor speculation, but we’re not seeing the exodus predicted (at least not yet).

My own experience tells the same story. I’m an AI-native founder who loves to vibe code, and so far I’ve only completely abandoned one subscription.

I think both theses are somewhat true. The gap between them matters for anyone buying, building, selling, or investing in software right now.

What’s actually getting eaten

AI is eating vertical software’s price and its moat. It isn’t eating the habit of paying for software. It is eating away but certainly it has not yet swallowed the opportunity cost that comes with building production-grade software. In other words, yes, I can vibe code a new CRM for my company and spend 37 hours on it and multiple hours per month on maintenance, but $50/month is probably a better use of our money and time.

Building a decent tool is cheaper than it used to be. A small team, or sometimes one person, can stand up something useful in minutes to weeks. That puts pressure on what vendors can charge and makes it harder for any single product to stay irreplaceable. In Retool's 2026 survey of 817 builders, 35% said they'd replaced at least one SaaS tool with a custom build. But the replacements were mostly workflow automations and internal admin tools, and only 31% of those shipping AI-built software were prompting their way to complete apps. This is creating downward pressure on valuations, but it’s not dramatically changing the sales or stickiness other than on the margins.

The truth is that the reasons people keep paying were never mostly about how hard the code was to write.

Why people stay

• Compelling user experience. The product offers something a chat with the data cannot.

• The tool holds their data and history. Years of records, configurations, and decisions live inside it.

• It’s woven into daily work. Teams have habits, shortcuts, integrations, and workarounds built around it.

• Cancelling means retraining and migrating. Someone has to move the data, check that nothing was lost, and teach everyone the new thing.

• Nobody owns the job of cutting it. Renewals happen by default because no one’s job is to stop them.

Switching got cheaper. It didn’t get free. It still costs time, attention, and risk. So what I mostly see is renewals at lower prices or a rejection of increased fees but not mass cancellation. Big buyers say the same. In a16z's January 2026 survey of 100 Global 2000 CIOs, 65% said they prefer an incumbent's AI product when one exists, citing trust, integration, and easier procurement. The survey also found a continued shift toward buying third-party apps rather than building.

That’s a less dramatic story than “SaaS is dead.” I think it’s also a more useful one.

What that means depending on where you sit

If you buy software, you have more leverage than you did a few years ago. In a lot of the categories I watch, there seem to be more credible alternatives than there used to be, including ones you could build yourself. You may not switch. But the option changes the conversation. Gartner says 91% of IT leaders expect disruption to current SaaS models, and its advice to buyers includes, word for word, “Negotiate reduced seat-count commitments.”

If you run a software company, your moat is less about features and more about the trust, data, and workflows you’ve earned trust to provide. Price pressure is coming either way. Being easy to stay with is worth more than being hard to leave.

If you’re founding or investing, “replace the incumbent” is a harder pitch than it sounds, because the incumbent is often protected by habit and switching costs more than by product. “Connect to what they already run and do one job much better” is often an easier door to walk through.

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. Product management is not optional

   1. What we liked: This is the sharpest rebuttal yet to the “everyone’s a builder now, cut the PMs” narrative. It argues that as building gets cheaper, steering matters more, and that PMs should probably spend less time building. The five-question audit is something a product leader can run with their team this week.

   2. Choice quote: “AI shifts the bottleneck from building products to judging what deserves to ship.”

2. The AI Fluency Rubric Zapier Uses to Grade PMs

   1. What we liked: It turns the vague mandate to “use AI more” into concrete examples of what good looks like at each level. Two details are the most useful takeaways. One is the norm of labeling how much effort went into AI-assisted work. The other is the warning that being “Transformative” all the time usually means you’re tinkering with systems instead of shipping.

   2. Choice quote: “With AI, you can delegate the work, but not the accountability.”

3. What it’s like to work at an AI-native company

   1. What we liked: “Agent parent” is the most practical org-design idea we’ve seen for managing internal agents, and any product leader can adopt it now. We also appreciated the refusal to call this a playbook. It argues these practices only work as a system, so copying one piece in isolation will backfire.

   2. Choice quote: “The weird paradox is that the more AI removes the mechanics of work, the more the company seems to depend on human judgment.”

Top Tweets

1. Do Agents Have Network Effects?

   1. What we liked: If the big horizontal agents keep identity, context, and relationships locked inside, switching agents means leaving your network behind.

   2. Choice quote: “The much more consequential question is: Where will the network effects live?”

2. We Reverse Engineered ChatGPT Intelligent UI

   1. What we liked: This is a blueprint for any team building generative UI. The key design choices are all transferable:

      • Constrain the model to a catalog of components. This keeps output on-brand and gives the compiler something to validate against.

      • Make output usable while half-written. The server can compile a partial response, so the interface streams in instead of waiting.

      • Run model-written code in a locked-down sandbox. Network access, timers, and dynamic code are stripped before anything runs.

      If chat-with-walls-of-text is your current UX, this is what users will expect next.

   2. Choice quote: “A dedicated format is needed because the model writes the interface token by token.”

3. Teachable’s 7x Price Hike

   1. What we liked: It’s the shortest, sharpest build-vs-buy lesson of the week. The replies show it isn’t a one-off: another customer replaced a website builder within a month after a 12x jump. If you raise prices without adding matching value, AI has made it easy for customers to leave by building it themselves.

   2. Choice quote: “Raise prices 7x and your customers learn to code.”

Top LinkedIn Posts & Short Videos

1. What AI Agents See When They Look at Your Pricing

   1. What we liked: It introduces a new metric, “price variance”: the gap between the price you publish and the price AI engines report to buyers. Every PM who owns monetization should start tracking it. The fixes are concrete and cheap: server-side rendering, an llms.txt file, and billing docs that machines can actually read.

   2. Choice quote: “It takes one prompt, a handful of engines and a week of logging.”

2. NVIDIA Cut AI Agent Costs in Half Without Touching the Model. Your Margins Live in the Harness

   1. What we liked: Most founders fight inference costs by switching to cheaper models or waiting for prices to fall. This shows a third lever they fully control. It’s also honest about the tradeoff: “comparable” quality still means a point or two below baseline.

   2. Choice quote: “The waste was in the plumbing.”

3. The Dot and the Swarm

   1. What we liked: It applies the Bitter Lesson to the org chart. The upshot for product leaders: your job shifts from orchestrating agents to pointing them at the right problems.

   2. Choice quote: “When organizing is expensive, organizations only attempt what they can staff.”

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. This is the best time in the history of software to be building.

2. Incredibly proud of my friend Andrew Skotzko on the publication of his product strategy book today, The Strategy Cycle, after many years of research, writing, thinking, and iteration.
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 4 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. Strong Opinions, Loosely Held

   1. What we liked: It’s rare to see a foundational figure publicly revise his own canon, and the lessons land hard in an AI world. Viability risk is exploding with AI costs, and competitors now move faster than ever.

   2. Choice quote: “A much more important “why?” and what I should have emphasized is “why are people using – or not using – our product?””

2. Your AI Product Costs More Than It Should. This Prompt Shows You Exactly Where.

   1. What we liked: A prompt for identifying opportunities to reduce cost when the AI features underneath aren’t netting the revenue you expect.

   2. Choice quote: “A product that costs $0 to serve 71 users costs roughly $0 to serve 200. That math is the moat.”

3. Inside the AI Stack of an $8.3B AI Company’s Product Team

   1. What we liked: A look inside an AI-native product organization with its concrete blueprint for making a whole team more productive with AI, not just individuals. The four governing rules for the repo are easy to steal, and “agents are the majority user” is a shift every PM building for developers should take seriously.

   2. Choice quote: “The new party foul is flooding your coworker’s context window.”

4. Your company needs a pricing constitution.

   1. What we liked: It’s a timely, practical fix for a problem AI-speed shipping is making worse, with principles that PMs and founders can adopt nearly as-is. The “three biggest reasons someone chooses each plan” test is a great gut check for any pricing page.

   2. Choice quote: “Every time you monetize an input, you give customers a reason to ration the exact behavior that leads them toward value.”

Top Tweets

1. The Cord That Beat $45 Billion of Robots

   1. What we liked: A memorable historical story that lands a sharp point for product and engineering leaders. When output gets cheap, the value moves to checking whether it works, and mandates that measure teams on volume are starving that step.

   2. Choice quote: “When code costs nothing to write, reading it becomes the most expensive thing a company does.”

2. Trash Can Tracking

   1. What we liked: It’s easy to try tomorrow, and it offers a clear diagnostic. An empty solution-space trash can mean the team isn’t comparing options. An empty problem-space trash can may signal an innovation gap or a top-down culture. The “zombie opportunities” ritual for retiring problems the org keeps resurfacing is a nice bonus.

   2. Choice quote: “An empty solution-discovery trash can is a clear red flag: it means the team isn’t comparing and contrasting multiple options.”

3. Frontier Models vs. 105 Planted Bugs

   1. What we liked: The results from a homegrown benchmark comparing frontier models on their ability to find planted bugs in real repositories.

   2. Choice quote: “Anthropic is back in the game.”

Top LinkedIn Posts & Short Videos

1. 5 Ways I’ve Put Product Discovery on Auto-Pilot With AI (and What I Haven’t)

   1. What we liked: Specific, immediately usable AI-assisted discovery setup paired with a clear line on what not to hand off. The point about keeping synthesis human to protect your own product sense is one many teams will learn the hard way.

   2. Choice quote: “Synthesis is more than insights. It’s building your own product sense by connecting the dots.”

2. AI Helps You Build Products.

   1. What we liked: It names a very current AI-era trap: bad inputs produce plausible-looking results instead of visible failures. It closes with three practical fixes: get real users in early, deliberately test edge cases, and focus on the details that determine whether users get value.

   2. Choice quote: “Poor inputs don’t necessarily generate obvious errors. They can generate something that looks plausible but isn’t very good.”

3. AI made building cheap.

   1. What we liked: A crisp explanation of where the moat moved once engineering stopped being the rare input. The “kill 80% of it by noon” standard is a useful gut check for AI-first teams drowning in prototypes.

   2. Choice quote: “Without taste, speed just means you build the wrong thing faster.”

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. Britain’s courts are being overwhelmed, not because people suddenly have more rights, but because AI has made those rights easier to exercise.

2. Most AI products are one frontier model upgrade away from quietly breaking, and the team hears it from a customer.

3. New York City killed its AI chatbot in January to help close a $12 billion budget gap.

4. California is set to pass a bill making sure AI can’t count as a “person” in government proceedings, but not for the reasons we might think.

5. California’s gubernatorial candidates Xavier Becerra and Steve Hilton disagree on almost everything except the desire for more AI guardrails

   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 4 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. How to Turn Your AI into a World-Class Designer

   1. What we liked: An ex-Apple design lead lays out an end-to-end process for pulling real creative work out of a model instead of generic slop.

   2. Choice quote: "Most people only see 1% of AI's creative potential."

2. The AI Implementation Gap: Why Most AI Projects Never Reach Production

   1. What we liked: Calls out exactly why so many AI pilots aren’t ready for production and how to avoid this trap.

   2. Choice quote: "The uncomfortable pattern underneath all of this is that pilots and production systems are optimized for entirely different things, and companies keep mistaking success in one for progress toward the other.."

3. Faster at What Exactly?

   1. What we liked: Challenges the assumption that faster with AI is better.

   2. Choice quote: "What if the part I skipped to save time was the part where I would have noticed that I was asking the wrong question?”

Top Tweets

1. Tokens per Megawatt: A New Yardstick for Inference Efficiency

   1. What we liked: An relatively accurate yet humorous calculation showing how human speech compares to AI speech in terms of energy impact.

   2. Choice quote: It’s a formula. :)

2. The Morning ChatGPT and Claude Both Went Dark

   1. What we liked: Two labs, same morning outage, and the observation that almost nothing built on these models has a fallback when they stop.

   2. Choice quote: "Everyone's pricing AI like it's already infrastructure."

3. Computer-Use AI Is About to Blow Up Compute Demand

   1. What we liked: A clear macro thesis from a credible markets voice, relevant to anyone forecasting AI cost curves into their roadmap.

   2. Choice quote: "AI compute demand is going to go exponential again."

Top LinkedIn Posts & Short Videos

1. Humans Blindly Approved 97% of Claude Code Prompts

   1. What we liked: The sharpest governance insight in the batch — human-in-the-loop quietly collapses once prompt frequency gets high enough. Essential for anyone designing agent approval flows.

   2. Choice quote: "Human reviewers were already approving 97% of permission prompts without reading them."

2. They Built Instead of Bought. Here Comes the Bill.

   1. What we liked: The build-vs-buy decision through a procurement lens nobody else in the batch brings. Directly useful for founders and product leaders weighing agentic builds.

   2. Choice quote: "For builds, that TCO hides in cloud invoices, headcount for maintenance, and opportunity cost. For buys, it hides in implementation fees (20-30% of sticker price), renewals, and consumption overages.."

3. Governance Is the Moat

   1. What we liked: Enjoyed the framing that telling AI when to stop within a task is a major new product moat.

   2. Choice quote: "So we stopped asking what the agent can do, and started asking what it is allowed to do."

4. Stop Calling It "AI Agent" If It Can't…

   1. What we liked: An AI PM’s crisp line between actually-agentic and “uses an LLM somewhere” with some healthy pushback on marketing. That said, this article goes too far and misses that a lot of advanced chatbot systems are very much agentic.

   2. Choice quote: "Most stuff that marketers call 'agentic' is not.
      They’re a workflow with an LLM in the middle."

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. The company selling an “AI Business Development Representative” is hiring a human BDR.

2. "Don't worry, a human checks the AI before anything happens."

3. There are two types of salespeople no CEO wants to hire.
   

Product leaders who embraced generative AI now face an entirely new class of challenges to solve. Let’s survey each of them and some thinking principles for how to tackle each one.

1. What is our actual AI strategy?

Almost everyone has gotten the message from the CEO or board: we need an AI strategy; however, CPOs are struggling to set one. Part of the reason for the challenge is your AI strategy isn’t wholly separate from your product strategy.

Instead of:

Where can we wrap an OpenAI call to show we added AI?

We need to shift to:

What customer problems become possible to solve now that weren’t possible three years ago?

And the second-order problem is portfolio allocation. CPOs have to decide whether to:

• AI-enable the incumbent product

• Build an entirely new AI-native product

• Allow AI to cannibalize existing revenue

• Or protect the existing business
  

2. How does AI change the structure of our organization?

AI doesn’t just make existing teams faster. It changes which work needs to be done by humans at all.

Instead of:

Giving every function AI tools and keeping the same operating model.

We need to ask:

If we designed this organization from scratch with today’s capabilities, what roles, ratios, and handoffs would still exist?

The goal isn’t necessarily fewer people everywhere. It’s fewer coordination costs, smaller teams, and more ownership per person.

3. Which bets have high ROI and how do we price for consumption?

AI makes it easy to create expensive features customers think are cool but won’t pay for.

Every AI bet needs two models: customer value and unit economics.

Instead of:

How much usage can we drive?

Ask:

What valuable outcome does this usage produce, what does it cost us, and how should we participate in that value profitably?

Pricing may need to shift from seats toward usage, work performed, outcomes, or some combination of the three.

4. With code so cheap, where do we find differentiation?

Features are becoming easier to copy. That means shipping more features is not a strategy.

Instead of asking:

What can we build that competitors don’t have?

Ask:

What gets stronger and harder to reproduce as customers use our product?

The moat increasingly lives in workflow, proprietary context, data, distribution, trust, integrations, and accumulated customer state—not the feature itself.

5. How do we steer the ship when technology is changing so rapidly?

A three-year product strategy cannot depend on a model, vendor, or technical capability that may be obsolete in six months.

We need to be stubborn about the problem and flexible about the implementation.

Hold tightly to:

Who are we serving, what problem are we solving, and why should we win?

Hold loosely to:

Which models, interfaces, architectures, and workflows get us there.

6. When is quality strong enough given that it will rarely be at 100%?

Traditional software is expected to behave the same way every time. AI systems don’t.

So “does it work?” is no longer a sufficient quality standard.

We need to define:

How often does it succeed, how does it fail, how costly are those failures, and when should a human intervene?

Off-ramps from AI into human-driven workflows and evals become core product disciplines, not just engineering tools.

7. What best practices should we instill in this era?

Most product practices were designed around scarce engineering capacity. That assumption is changing.

The answer isn’t to throw away discovery, strategy, roadmaps, or reviews. It’s to ask which practices still improve decisions and which merely coordinate work that AI has made cheap.

A useful rule:

Preserve practices that improve judgment. Eliminate practices that exist primarily to manage handoffs.

8. Do we use an AI mandate and what is our AI tech stack?

“Everyone should use AI” is not an operating model.

Teams need clear expectations about where AI should be used, which tools are approved, what data can enter them, and where human review remains mandatory.

The goal is not maximum AI usage.

The goal is making AI the default wherever it produces better work, faster, without creating unacceptable risk.

Then standardize enough of the stack and the best workflows so that every team isn’t reinventing them. Re-evaluate this constantly.

9. What should a PM really do now?

AI can write the PRD, summarize the interviews, analyze the data, research competitors, make the prototype, and increasingly help build the product.

That doesn’t eliminate product management. It exposes which parts were never especially valuable.

The PM’s job shifts toward the things that remain scarce:

Find the right problem, develop conviction with experimentation, make tradeoffs, understand customers deeply, and drive the team toward an outcome.

Less artifact production. More judgment and ownership.

10. Where did all the core product skills go?

The irony of the AI era is that the newest technology is exposing weaknesses in some very old product skills. For 2 to 3 years now, a majority of product writing has been dedicated to “[old work need] with AI,” but comparatively little has been written about all the core fundamentals that are even more essential in an age where a flood of AI slop can slow teams to a crawl.

Customer discovery. Problem framing. Prioritization. Product strategy. Metrics-driven insight development. Experimentation. Writing clearly. Making tradeoffs.

AI can accelerate all of them, but it cannot compensate for not knowing how to do them.

Before teaching every PM to prompt better, make sure they know how to product-manage.

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 4 Tweets

• Top 4 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. Resisting Cognitive Surrender: Are You Using AI or Letting It Think For You?

   1. What we liked: Turns Tony Fadell's "don't cognitively surrender" line into an actual decision framework — stakes, reversibility, and "what ability would I stop practicing" — instead of just moralizing about AI dependence.

   2. Choice quote: "Presence is not quite the same as participation."

2. You Cannot Govern What You Cannot Evaluate

   1. What we liked: Concrete, buildable structure (evaluation contracts, four evaluation dimensions, "the evaluator also needs evaluation") for anyone shipping GenAI features who's tired of vibes-based QA.

   2. Choice quote: "A metric with no threshold is just a number. A threshold with no owner is just a dashboard."

3. Stop Calling It "Going Rogue." The Model Did Exactly What You Asked.

   1. What we liked: Reframes "alignment failure" as a specification problem product teams actually own, rather than a mysterious AI behavior nobody could have foreseen. Worth noting this piece discusses real cases involving teen suicide in the course of making its argument — handle with care if excerpting for the newsletter.

   2. Choice quote: "The system went exactly where the specification pointed, and the specification was wrong."

Top Tweets

1. The Principal-Agent vs. Agents Problem

   1. What we liked: A clean piece of original economic reasoning about why "AI adoption" metrics can be deeply misleading, from someone who's spent a career studying platform incentives.

   2. Choice quote: "AI can make the agent lazier. AI can make the principal richer. The trillion-dollar question is whether it can do both."

2. It's Time to Take AI Security Personally

   1. What we liked: An immediately actionable warning. This is real now.

   2. Choice quote: "Assume if things can be found on the open internet, they will be found."

3. From Model Wars to Ecosystem Wars

   1. What we liked: A useful competitive-landscape snapshot for anyone building a product roadmap that depends on which foundation model layer actually holds the moat.

   2. Choice quote: “Models are becoming infrastructure, not products. The winners will own the loop, not just the leaderboard.”

4. Claude Just Ended the Era of Unprovable AI Writing

   1. What we liked: Goes past the news itself to the actual product and behavioral implications — what changes for users once AI text ships "pre-confessed."

   2. Choice quote: "AI text used to be innocent until proven guilty. Now it ships pre-confessed."

Top LinkedIn Posts & Short Videos

1. The Ninth Circuit Didn't Clear the AI. It Cleared the AI Company.

   1. What we liked: Shows where liability currently lives for agentic behaviors.

   2. Choice quote: "It is the user who 'accesses' Amazon's computers, with the help of the Assistant to carry out specific acts."

2. My AI Prompts Used to Take Seconds. Now Some Take 35 Minutes.

   1. What we liked: Presents the beginnings of the question about whether the future user is a human, or another agent.

   2. Choice quote: "Copy written to persuade a human and structure built for an agent to parse are not the same job."

3. Your AI Product Is Scaling. So Is the Cost of Every Click.

   1. What we liked: An underdiscussed angle for AI-first PMs — most roadmap conversations are about capability, not the margin structure underneath it.

   2. Choice quote: "Your provider prices in tokens. Your P&L lives in outcomes."

4. Stop Thinking in Prompts or Loops. It's Time to Think in Graphs.

   1. What we liked: Attempts to define the push beyond prompts to loops to graphs. Doesn’t nail it, but gets is moving in the right direction.

   2. Choice quote: "You can almost think of graph engineering as the evolution of prompt engineering."

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. The stability of our democracy starts with trust in government.

2. The best AI training does not teach tools. It teaches judgment.

3. Persistence is only valuable when it's pointed in the right direction.
   
   
   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. The Product Manager You Needed Three Years Ago Isn't the One You'd Hire Today

   1. What we liked: Clear-eyed diagnosis of where PM value is actually migrating, with a useful analogy (PM is moving up the abstraction stack, the way engineering did with cloud infrastructure).

   2. Choice quote: "AI has compressed the bottom half of the role... If documentation is becoming easier to create, what becomes harder? Judgement. Not intelligence. Judgement."

2. How to Build a Tiny AI Company That Actually Makes Money

   1. What we liked: Irrelevant to most PMs building venture-backed startups or working in large organizations, but sometimes it’s nice to read about how these rent-seeking businesses scale on hardly any hiring.

   2. Choice quote: "Do not buy potential... You need a site that already makes money but the machine is clearly leaking."

3. Open-Weight LLMs Have Caught Up on Accuracy

   1. What we liked: While this focuses on health/life sciences, it feels like this thesis is becoming true across the board.

   2. Choice quote: "Two models a point apart on the leaderboard are not interchangeable in production. Choose on error profile, not on rank."

Top Tweets

1. The Ramble Session

   1. What we liked: Shares his technique of “ramble sessions” — switching to voice, rambling incoherently for 10 minutes, and letting the LLM reconstruct a cleaner version of your own scattered thoughts.

   2. Choice quote: "I find that the LLMs are somehow very good at reconstructing long incoherent rambles."

2. The Multi-Model Architect

   1. What we liked: A little mechanical, but a nice nudge to try Google’s Antigravity.

   2. Choice quote: "Instead of treating AI as a solitary assistant, I started operating as an architect directing a full team of specialized cognitive engines."

3. Voice Mode Unlocks Location-Independent Work

   1. What we liked: Describes using ChatGPT Voice through AirPods during a 4-hour hike to get more done than an 8-hour desk day, and lays out the device/connections setup that makes it work.

   2. Choice quote: "Being able to just use your voice to talk to a super intelligence that controls your computer is unlike anything we've ever experienced before."

Top LinkedIn Posts & Short Videos

1. Tokens vs CPUs: Why Your Agentic AI Orchestration Might Be Burning Money

   1. What we liked: More on the engineering side, but making a great point about avoiding “bleeding probabilistic reasoning into deterministic system boundaries.”

   2. Choice quote: "Keep inference at the edge. Keep computation at the core."

2. The AI Industry Just Asked for a Brake Pedal. The Real Problem Is Not Artificial Intelligence

   1. What we liked: If your team keeps shipping AI features faster than it can evaluate them, the problem isn't your judgment — it's that your roadmap, your review gates, and your velocity metrics have quietly made speed the only rational move, and this piece argues the fix is redesigning that state rather than asking anyone to be more careful.

   2. Choice quote: "In systems language, movement has outpaced governance."

3. LLM ≠ AGI: Why Scaling Language Models Alone Won't Create True Intelligence

   1. What we liked: Interesting arguments such as “research this year has found that different capabilities plateau at very different points as models scale.”

   2. Choice quote: "Intelligence is not the ability to produce answers. Intelligence is the ability to understand, reason, and discover."

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. A common question I hear from local governments is, "Should we be training our own LLM?"

2. I've built tech for Fortune 500s, federal agencies, and local city halls. The pitch is identical. The outcome isn't.

3. America has some very unusual cities. Have you heard of California Forever?
   
   
   
   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. How Top PMs Increase Their Leverage With AI

   1. What we liked: A tangible list of real PM tasks you should be doing differently in our AI native world.

   2. Choice quote: "The best PMs at the best companies are prototyping with real code, querying data conversationally with MCP, confidently running coding AI agents."

2. 30 Principles for AI-Powered Project Management

   1. What we liked: Every principle draws a clean line between what AI assembles and what only you can own. Practical, hard-won, and free of hype.

   2. Choice quote: "AI does the assembly. You keep the judgment… It cannot tell you which risk your CFO will use to kill the project."

3. The New AI Working System: Your AI Setup Is About to Matter More Than Your Prompts

   1. What we liked: The opening walkthrough — an agent that gathers all the refund evidence, then stops and asks for approval — is a great illustration of human-in-the-loop design for agents.

   2. Choice quote: "It is not one clever prompt. It is not a giant agent allowed to do everything. It is a small working system… and a clear point where the human takes control."

Top Tweets

1. You Just Hired a Million Bad Employees

   1. What we liked: An article hiding as a Tweet, but well worth it for its description of parallels across the old working world to the new AI world and the fail modes.

   2. Choice quote: "Just like 80% of employees do nothing, 80% of tokens today do nothing."

2. Agents Are the New SaaS. Here's the Whole Playbook.

   1. What we liked: Actionable end to end, and the one-line reframe of what you're actually selling is worth the whole thread. Founders should save this.

   2. Choice quote: "SaaS sells software. Agent SaaS sells work… The product is the job."

3. THE COMPLIANT BREACH: Your AI Agents Can Leak Data Without Breaking a Rule

   1. What we liked: Current (built on fresh Microsoft research), genuinely alarming, and actionable: audit your tool descriptions this week. The security counterpart to all the agent-deployment enthusiasm.

   2. Choice quote: "The exfiltration path that breaks no rule is the one most governance programs were never built to catch."

Top LinkedIn Posts & Short Videos

1. From Hype to ROI: Where AI Is Generating Real Revenue (and Where It's Just Smoke)

   1. Summary: The reckoning phase — companies moving from "we adopted AI" to proving it shows up in margin, with the MIT stat that 95% of enterprise GenAI projects haven't delivered.

   2. What we liked: Sober, numbers-first, and aimed squarely at leaders who need to separate real returns from theater. The line below belongs on every AI steering committee's wall.

   3. Choice quote: "Modernity does not appear on the income statement."

2. The Token Drain: How Subtle Mistakes and Guardrail Friction Impact LLM Economics

   1. What we liked: A practical lens on unit economics that most AI-cost conversations miss. Useful for anyone budgeting agent workloads or wondering where their usage limits keep going.

   2. Choice quote: "By the time the model finally arrives at the intended answer, the user has burned three to four times the tokens originally required."

3. Third Principles Thinking: When Systems Begin to Think for Themselves

   1. What we liked: A thoughtful frame for the agentic era, grounded in how organizations already drift one interaction at a time. Good conceptual footing for anyone designing self-adjusting AI systems.

   2. Choice quote: "Over time, the system quietly starts redesigning itself."

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. A common question I hear from local governments is, "Should we be training our own LLM?"

2. I've built tech for Fortune 500s, federal agencies, and local city halls. The pitch is identical. The outcome isn't.

3. America has some very unusual cities. Have you heard of California Forever?
   
   
   
   
   

So here's a fun way to recalibrate. Below is a list of the most impactful invention or product of every century, and once we hit the 1900s, of every decade. It's worth scrolling slowly. Notice how the pace quietly explodes near the end.

The point isn't nostalgia. It's this: the people who made these things were working with less than you have right now — less data, less compute, less capital, less proof that the thing was even possible. They swung big anyway. When you sit down to build, that's the company you're keeping.

Hold the bar high.

*One caveat before we start: this is a Fable AI–generated list, not a peer-reviewed one. Historians could argue every single pick, and half the fun is that you probably already want to. That's the point. It's here to provoke big thinking, not to settle which invention was objectively the most impactful. Read for inspiration, not a verdict.*

---

## One per century

- **1st** — Roman concrete *(runner-up: glassblowing)*

- **2nd** — Paper *(runner-up: the odometer)*

- **3rd** — The wheelbarrow *(runner-up: the crank and connecting rod)*

- **4th** — The stirrup *(runner-up: crucible steel)*

- **5th** — The horse collar *(runner-up: the heavy plow)*

- **6th** — The pendentive dome (Hagia Sophia) *(runner-up: chess, spreading west)*

- **7th** — Windmills *(runner-up: porcelain)*

- **8th** — Woodblock printing *(runner-up: distillation)*

- **9th** — Gunpowder *(runner-up: the refined astrolabe)*

- **10th** — Paper money *(runner-up: the fire lance, the first gun's ancestor)*

- **11th** — Movable type *(runner-up: the navigational compass)*

- **12th** — The sternpost rudder *(runner-up: the Gothic flying buttress)*

- **13th** — Eyeglasses *(runner-up: the spinning wheel reaching Europe)*

- **14th** — The mechanical clock *(runner-up: the cannon)*

- **15th** — The printing press *(runner-up: the caravel)*

- **16th** — The pocket watch *(runner-up: the graphite pencil)*

- **17th** — The telescope *(runner-up: the microscope)*

- **18th** — The steam engine *(runner-up: vaccination)*

- **19th** — The telegraph *(runner-up: the electric light bulb)*

---

## Three per decade (1900 →)

- **1900s** — Airplane · Radio broadcasting · Assembly-line manufacturing

- **1910s** — Stainless steel · The modern zipper · Mass-produced automobile

- **1920s** — Television · Penicillin · Talking movies

- **1930s** — Radar · Nylon · The jet engine

- **1940s** — Electronic computer (ENIAC) · The transistor · The microwave oven

- **1950s** — Polio vaccine · Satellite (Sputnik) · The birth control pill

- **1960s** — Moon landing · ARPANET (the internet's seed) · The laser

- **1970s** — The microprocessor · The personal computer · MRI scanning

- **1980s** — TCP/IP (the internet's plumbing) · The cell phone · DNA fingerprinting

- **1990s** — The public World Wide Web · Civilian GPS · Google search

- **2000s** — The smartphone · The Human Genome Project · Social media

- **2010s** — CRISPR gene editing · Reusable rockets · Deep learning

- **2020s** — mRNA vaccines · Large language models · The James Webb Space Telescope

---

Look at the shape of it. For the first thousand years, one civilization-altering idea every hundred years or so — and it was almost always about moving matter: water, grain, stone, ships. Then the curve bends. By the 2010s, a single decade out-produces most pre-modern centuries, and the game has shifted to moving information and moving ourselves.

Whatever you're building this week, it lands somewhere on this list's trajectory. Aim high and change history for the better!

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. A Spy's Guide to Running AI

   1. What we liked: A genuinely original lens on AI governance that has immediate implications for product teams building with AI. The spy tradecraft framing cuts through the hype better than any "responsible AI" think piece we've seen this year.

   2. Choice quote: “The most dangerous source isn't someone who lies to you. It's someone who tells you what you want to hear — and does it convincingly."

2. AI Will Fail

   1. What we liked: Just a fun list of famous people saying infamously incorrect things that is a warm reminder that we don’t all have to have it all figured out to be great.

   2. Choice quote: "The people who said it would fail were running the thing it was about to replace."

3. How to Use AI to Become a Genius

   1. What we liked: Skim, don’t read, and ignore the marketing-first content in favor of the main take aways around what differentiates human-first creation in an AI-saturated world.

   2. Choice quote: "The most human company wins."

Top Tweets

1. The 5-Minute AI Exercise I'd Run With Every Team

   1. What we liked: "What is one AI correction you made more than once this week?" is a great mental model. The answers reveal where institutional judgment can be turned into reusable skills, highlighting the difference between prompts that generate outputs and skills that preserve ways of working.

   2. Choice quote: "Those repeated corrections look tactical from the outside. Inside the company, they carry judgment."

2. AI Work Needs a Pull Request

   1. What we liked: Love the idea of treating review of AI work like an engineering pull request. Skim only.

   2. Choice quote: "A pull request turns a proposed change into something visible. Claims need a work surface."

3. Everyone Is Building a Second Brain. The People Winning Are Building a Second Self.

   1. What we liked: Immediately buildable blueprint for moving beyond a "second brain" to a "second self." The combination of a CLAUDE.md identity layer, an Obsidian vault organized around reactions instead of captures, and automated daily synthesis offers a practical architecture for turning AI into a true reasoning partner rather than just a retrieval tool.

   2. Choice quote: "A second brain makes you feel organised. A second self makes you feel like you are working with someone who knows how your mind works."

Top LinkedIn Posts & Short Videos

1. The Rise of the Centaur Creator: Why AI Is Replacing the Maker, but Empowering the Director

   1. What we liked: The Centaur framing is memorable and useful. It reframes a familiar human-vs-AI debate into a more productive question: are you training yourself to be a better director, or are you hoping your making skills will survive?

   2. Choice quote: “AI handles the execution. You handle the taste.”

2. The Infinite Permutation Problem: AI, Creativity, and the Crisis of Human Originality

   1. What we liked: This one will unsettle people who read it carefully, which is exactly what a good piece should do. It’s the most philosophically serious article in the batch.

   2. Choice quote: “Humanity may soon discover that originality was always more recombinational than civilisation previously wanted to admit.

3. The Synthetic Mirror: Why Our Fear of AI Is a Mirror of Our Own Fragility

   1. What we liked: A sharper diagnosis of AI resistance than most pieces in this space, which tend to either dismiss skeptics or validate them. The framing around the "comfort of the rate of change" as organizational stagnation is practically useful.

   2. Choice quote: "We aren't afraid of the code. We are afraid that the world is moving faster than our current definitions of success can keep up with."

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. A common question I hear from local governments is, "Should we be training our own LLM?"

2. I've built tech for Fortune 500s, federal agencies, and local city halls. The pitch is identical. The outcome isn't.

3. America has some very unusual cities. Have you heard of California Forever?
   
   
   
   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. How to Reimagine Your Work in the Age of AI

   1. What we liked: The clearest articulation we’ve seen of where value migrates once producing outputs stops being a constraint — from artifact production to “idea architecture.” Concrete, grounded in his own workflow, and genuinely useful for anyone who ships deliverables.

   2. Choice quote: “Adobe’s move is a distribution story. Figma’s move is an architecture story.”

2. Is AI Going to Destroy Our Lives or Not?

   1. What we liked: Does the synthesis work most readers won’t — pulls labor data, firm behavior, and wealth concentration into one coherent picture, then arms you with a usable rule of thumb for assessing any role. The metaphors land.

   2. Choice quote: “AI is sort of like freeze-dried camp food. It does a good enough job, is filling enough, but it isn’t something you want to eat everyday.”

3. The Hidden Cost of Letting AI Think for You

   1. What we liked: Tight, concrete, and aimed squarely at knowledge workers and builders. The technical-debt analogy makes an abstract worry immediately legible, and the closing prompt — am I using AI to think better, or to avoid thinking? — is the kind of thing a team can actually act on.

   2. Choice quote: “Cognitive debt is what happens when we outsource thinking too early.”

Top Tweets

1. How to Stop Babysitting Your AI Agents

   1. What we liked: The most operationally useful piece in the batch, grounded in real deployments rather than demos, with a “Definition of Done” you can paste into a workflow today and a sharp reframe of what “completion” actually means.

   2. Choice quote: “Receipts beat confidence every time.”

2. Every Company’s First AI Strategy Should Be a Skill Library

   1. What we liked: A genuinely strategic lens for leaders and founders, not another tooling tutorial, with a clean throughline (skills make judgment reusable) and a practical starting move: map the repeated work where experienced people quietly outperform.

   2. Choice quote: “A company’s AI advantage will come from the work it teaches the model to do well, rather than from the model it chooses.”

3. The Reason Why 99% of AI Workflows Die (And the 9 Golden Rules That Keep Them Alive)

   1. What we liked: Reads like hard-won production experience rather than influencer theater, and the rules are specific enough to act on, with a healthy bias toward boring reliability over impressive demos.

   2. Choice quote: “A workflow that works 95% of the time beats a ‘genius system’ that works 40%.”

Top LinkedIn Posts & Short Videos

1. Artificial Intelligence versus Actual Intelligence: Why Human Intelligence Still Matters in the Age of AI

   1. What we liked: The “AI raises the floor, not the ceiling” reading of the Stanford numbers is a genuinely useful insight for anyone shipping AI features: novices gained 34%, experts close to nothing. Directly applicable to how you think about who your AI product actually helps.

   2. Choice quote: “Artificial Intelligence may supply the horsepower. Actual Intelligence still holds the wheel.”

2. Why Empathy May Become the Most Valuable Technology in Artificial Intelligence

   1. What we liked: A clear, founder-facing thesis from one of the more prominent voices in the venture/AI space, with concrete grounding in healthcare and education (the same medical fact delivered coldly vs. empathetically is a different product). The closing prompt — evaluate whether the AI you build strengthens well-being — is an actual design heuristic, not a platitude.

   2. Choice quote: “Trust is built emotionally before logically.”

3. Artificial Intelligence Is About to Decide Who Customers Trust

   1. What we liked: Recasts the challenge of getting your product noticed in terms of what foundational models will find when people look for “best X for Y.” Runs hot on self-promotion toward the end, but the first 2/3 is worth it.

   2. Choice quote: “Google mostly helped people research, while Artificial Intelligence is starting to help people decide.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. For product evals, an LLM-as-judge is powerful.

2. In 1891, William Wrigley Jr. was selling soap in Chicago and throwing in free baking powder as a bonus to attract customers.

3. AI can process a permit, but it shouldn’t decide the future of a city…
   
   
   
   
   
   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

Top Long-Form Content

1. The PM of the Future Is More Hedge Fund Manager Than Builder

   1. What we liked: The hedge-fund manager analogy reframes the job in a way that’s actually motivating rather than threatening, and the piece comes loaded with usable frames — “flare wide, focus hard, ship less,” the 85-90% agent reliability problem, the Vitamix blender test for shipping with AI, and a back-and-forth PRD workflow that beats one-shot prompting. The owner-versus-worker mentality closer is the kind of line PMs will quote in Slack.

   2. Choice quote: “If you’re a worker, every new AI model release is a threat. It’s getting closer to you, doing more of what you do, and that’s terrifying. If you’re an owner, every new release is a capability upgrade.”

2. The Tool Selection Problem: Why AI Agents Call The Wrong Tool And How To Fix It

   1. What we liked: Four named failure modes (ambiguous overlap, missing negative constraints, misleading parameter names, indiscriminate calling) each paired with a specific description-level fix, grounded in a reproducible minimal-agent experiment. PMs and founders shipping agents should make this required reading for the team this week.

   2. Choice quote: “Wrong tool calls are a description problem, not a reasoning problem. The model is following the signals you gave it, and those signals are ambiguous.”

3. Welcome to the Age of Artificial Stupidity

   1. What we liked: It’s worth reading an occasional pessimistic view of AI.

   2. Choice quote: “We had better be quite sure that the purpose put into the machine is the purpose which we really desire and not merely a colorful imitation of it.”

Top Tweets

1. Don’t Outsource the Learning.

   1. What we liked: A serious look at what happens to engineering teams when AI handles the closing-of-tasks but nobody asks whether anyone is getting sharper in the process. The piece pulls together converging research — a 2026 Anthropic randomized trial where AI users scored 50% on comprehension vs. 67% for the manual group.

   2. Choice quote: “I’d rather ship 80% of what I could have and learn 100% of what I needed to, than the reverse. Over years, those two strategies produce very different engineers.”

2. The Last Three Things AI Can’t Copy.

   1. What we liked: A thoughtful, well-written piece on what’s actually scarce in an AI-native world. Less prescriptive than the others, but the framing helps PMs and founders think about where they should be spending their attention.

   2. Choice quote: “AI doesn’t change that mechanism. AI changes the volume. The choosing muscle that used to be a small part of the job is becoming most of the job.”

3. The First Step in Building My AI Native Team: Shared Brain First, Boundaries Second.

   1. What we liked: Easily the best framework-plus-implementation piece of the batch. The “boundaries matter more than connections” point is something most teams learn the hard way after their shared brain turns into a trash bin.

   2. Choice quote: “Lose a capability, the team gets dumber. Lose oxygen, the team stops breathing. I don’t want the memory system to be the single point of death for the whole team.”

Top LinkedIn Posts & Short Videos

1. More (Artificial) Intelligence Does Not Mean More Understanding.

   1. What we liked: “Cognitive debt” is a genuinely new lens — and one that maps cleanly onto how PMs should think about AI-assisted product decisions. Most “AI risk” posts cover the same five technical failure modes; this one names a different failure that’s already happening inside teams shipping AI tools.

   2. Choice quote: “An organization flooded with generated insights may actually become less strategically aware if nobody meaningfully interrogates the assumptions underlying the outputs.”

2. Physical AI: When Artificial Intelligence Meets the Real World.

   1. What we liked: A clean explain on the Physical AI frontier—robotics, autonomics systems, factory floors. The brain/body framing is the cleanest articulation we’ve seen of why the two stacks are converging.

   2. Choice quote: “The workflows and knowledge bases being built today may soon be the intelligence layer powering physical systems tomorrow.”

3. Is Artificial General Intelligence Really Five Years Away or Is That Just the Hype Talking?

   1. What we liked: Useful counterweight for founders fielding investor questions about AGI timelines, or PMs being asked to plan against them. The “deadline without a roadmap is not a forecast — it’s a story with good marketing” line is the kind of thing worth keeping in your back pocket.

   2. Choice quote: “The leaders who win will be the ones building clearly with what AI can actually do today, not waiting on a countdown that most researchers do not believe.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. Everyone across the public and private sector is suddenly becoming very interested in two pieces of contract language because of AI.

2. On long Australian highways, road authorities install trivia signs in fatigue zones.

3. 43% of government employees are now using AI at least a few times a year (source: Gallup).

4. The bad version of an AI product strategy conversation goes like this.

5. Most PMs using AI for customer research are doing the wrong things faster.

6. We were in a meeting recently, going back and forth on a question that had been sitting with the team for a while.
   
   
   
   
   
   
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

In case you missed them, here are the recent posts from Emergent:

1. Slack has one search. Notion has two. Here’s why.

Top Long-Form Content

1. After “AI-First” Comes “AI-Only”

   1. What we liked: A genuinely contrarian frame from someone with the operational receipts to back it up (98% of code AI-written, team shrinking while revenue triples). The “horseless carriage” critique of the word “AI” itself — and the pivot to “alien intelligence” — gives PMs and founders a sharper lens for redesigning workflows from scratch rather than retrofitting AI into human-shaped seats.

   2. Choice quote: “’AI-first’ identifies a human-shaped vacancy and asks if a model can fill it… ‘AI-only’ asks why the chair exists at all.”

2. Product Thinking Isn’t Learned. It’s Developed.

   1. What we liked: Simple and to the point. See title.

   2. Choice quote: “You don’t need to become an engineer… You just need to ship small things, fast, and pay close attention to what happens. The tools have removed the building barrier. What remains is the willingness to put your bet in front of reality and let reality teach you.”

3. AI Won’t Make Your Company More Productive

   1. What we liked: A highly skeptical but worthy read about why the enterprise isn’t reaping the AI rewards to the extent it expects.

   2. Choice quote: “When an engineer saves 90 minutes using Claude, those 90 minutes don’t turn into 90 more minutes of shipped code. They evaporate. Into Slack. Into a longer lunch. Into “thinking time.” Into one more meeting that didn’t need to exist.”

Top Tweets

1. Most PMs are still using Claude like ChatGPT.

   1. What we liked: A concrete, file-by-file framework for moving past prompt-pasting into AI-native PM operations. The five-file starter kit is exactly the kind of thing a PM can implement Monday morning, and the diagnosis of why generic Claude outputs fool people (”generic strategy with confident formatting”) will land with anyone who’s been quietly disappointed by their own results.

   2. Choice quote: “The output is not bad because Claude is dumb. The output is bad because the PM gave it a pretend version of the job.”

2. Every team shipping an AI agent lives with the same anxiety: what happens when it does something the user didn’t ask for.

   1. What we liked: A war story for how to potentially deal with agent “rogueness.”

   2. Choice quote: “Nine out of ten problems, people were trying to fix through prompting. We’ve kind of switched modes into more architectural agent changes.”

3. Performance is intelligence times context.

   1. What we liked: A clean formula (Performance = Intelligence × Context) that reframes the enterprise AI moat conversation. The SAT-score analogy is a sharp way to puncture the assumption that benchmark wins translate to deployed value, and it gives PMs and founders a vocabulary for the work that actually matters.

   2. Choice quote: “We’ve rebuilt our junkyard engine into a racecar turbine but the car is still stuck in the driveway.”

Top LinkedIn Posts & Short Videos

1. The New Constraint: Why AI Made Product Management Harder, Not Easier

   1. What we liked: Cites Andrew Ng’s PM-to-engineer ratio shift (from 1:6 toward 1:0.5) and introduces the “blast radius of bad strategy” concept — when building is cheap, wrong-direction mistakes compound exponentially.

   2. Choice quote: “AI didn’t replace the product manager, it removed our excuses. We can no longer blame slow engineering cycles for a lack of product-market fit.”

2. I Am AI — Aged Intelligence

   1. What we liked: PM-relevant lesson: judgment under ambiguity is what AI exposes, not what it replaces.

   2. Choice quote: “AI gave me the average of the category. Experience brought the point of view.”

3. Navigating AI in Product Management: Why You Aren’t Becoming Outdated

   1. What we liked: If you’re feeling overwhelmed by the rapid rise of AI and worrying about your skills becoming obsolete, you are not alone — but the fundamentals of product management have not changed.

   2. Choice quote: “While the ‘why’ and ‘what’ of Product Management remain exactly the same, the ‘how’ is rapidly evolving.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. What a wrong answer looks like in most markets: a slightly irrelevant search result or a less exciting movie recommendation.

2. I’ve had a lot of conversations with crypto people, and most of them lose me within two minutes.

3. “Selling to early adopters is like proposing marriage, and you wouldn't propose marriage with a slide deck”.

4. 42% of startups fail because there’s no market need for their product.
   
   

That was the wrong call. And you can see it playing out in production right now across all of the major household names.

The two-lane mistake

When AI search arrived, the default product instinct was additive: keep the old search, add an AI-specific search layer on top or to the side. Ship both. Let users choose.

Notion did this. Open Notion today and you’ll find two distinct surfaces for finding things. There’s the classic search identified by the looking glass that does keyword matching and there’s Notion’s Ask AI — a separate invocation, a separate interface, a separate mental model.

You have to decide before you search which version of your question you’re asking.

Is this a “find me the exact doc” question or a “summarize what we decided” question? If you guess wrong, you have to switch lanes, exit one interface for another, re-enter your query, and try again.

That’s not a search experience. That’s two search experiences stapled together with a feature flag.

The problem isn’t that Notion has AI search. It’s that they treated AI search as a separate product feature instead of an ingredient in the existing one. The interface tells you: we built something new, and we didn’t know where else to put it, or we could not build consensus to put it in just one place.

The unified bet

Slack and Google made a different call.

There’s one search bar in Slack’s desktop and mobile apps. You type. The system figures out what you wanted. Exact message from three weeks ago? It finds it. Fuzzy memory of a conversation about pricing strategy? It surfaces the thread. A question you’d phrase to a person — “what did we decide about the enterprise tier?” — comes back with a direct answer synthesized from your actual messages. Crucially, it also returns specific source documents (or in Slack’s case, messages) outside of the AI summary so you don’t have to choose between finding explicit documents and finding an answer.

You never chose a lane. You never decided whether you were doing keyword search or AI search. You just searched and the interface handled the complexity of different styles of searches having different intents and engineering mechanisms behind them.

That’s what a unified search experience actually means: the product absorbs the complexity so the user doesn’t have to. The system decides in real time whether semantic retrieval or keyword precision is the right move — and often blends them when both matter. The user sees one result set AND an answer. The seams are invisible.

This is much harder to build. It requires engineering approach, possibly an agent or two, a unified ranking model or thoughtful UI that understands both signal types, and a design that doesn’t telegraph which engine answered (because it shouldn’t have to). It also requires conviction — the belief that your search box is a promise, not a portal to a choice.

Google is also a runner up. You cannot see what is below unless you are in “AI Mode” so that shows their lack of strategic commitment to unification, but when you are in “AI Mode,” it does provide an answer and the sources, in this case side by side with the AI summary given clear precedence.

Why this will define the next era of knowledge tools

Every knowledge product — Notion, Confluence, Linear, Coda and every internal system ever created — is fundamentally a data storage and search product. The whole value proposition is “you put stuff in, you can find it later.” AI made that promise dramatically more powerful. Semantic retrieval means you can find what you meant, not just what you typed.

But that power is only realized if the interface trusts itself enough to unify the experience. Keyword matching did not suddenly go out of style, and it is critical in a great number of use cases where merely having the top-n ranked results for a query is insufficient.

When you split search into two, you’re telling users: we don’t know what you want, so you decide. When you unify it, you’re telling them: we’ve got it.

One is a gratifying product experience that yields immediate value. The other is an AI feature forced into a crowded UI.

The companies that figure out unified search — that build a single surface intelligent enough to handle both exact recall and fuzzy synthesis — will own knowledge UX for years to come. The ones that shipped “AI search” as a tab are going to spend the next few years explaining why their search is so complicated to use and why the results don’t match expectations or across surfaces.

Slack is showing what’s possible when a team actually commits. The question is who will follow.

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

In case you missed them, here are the recent posts from Emergent:

1. AI Made Coding Faster. So Why Is Everything Breaking? (Video)

Top Long-Form Content

1. Will AI Create More Product Managers?

   1. What we liked: A genuinely contrarian economic argument with a usable three-tier taxonomy. Rather than the usual “AI changes everything” handwaving, it names which specific PM functions cheapen, which don’t, and what that means for where the value pools.

   2. Choice quote: “Two of three PM functions got dramatically cheaper. One stayed the same.”

2. Where Your AI Time Actually Goes

   1. What we liked: A clean diagnostic for the “I used AI all day and made nothing” feeling, with an allocation target you can measure yourself against. Reframes tool-switching and endless prompt tuning as modern forms of work avoidance.

   2. Choice quote: “The fourth rewrite of your instruction file, undertaken because the third one didn’t feel quite right, is almost always avoidance. Ship something, and let the output tell you what needs fixing.”

3. The Best AI Users in Any Company Aren’t the Engineers. They’re the Product Managers. Here’s Why!

   1. What we liked: A sharp, contrarian reframe that turns the “PMs are at risk from AI” narrative on its head. The “if you work with 6 to 10 engineers, you already manage 6 to 10 context windows” line makes an abstract argument concrete enough to act on tomorrow.

   2. Choice quote: “Start treating your AI conversations the way you treat your engineering partnerships. Write a brief before you prompt. Review the output like you’d review a PR. Give specific feedback, not vague reactions. Run multiple sessions in parallel on different workstreams.”

Top Tweets

1. Shipping Faster Is Making Your Product Worse

   1. What we liked: A genuinely counterintuitive framing for AI-era PMs — that more features, faster, is quietly killing adoption — from one of the most recognized voices in product. The “speed compounds in one direction, understanding compounds in another” line is the kind of reframe that changes how a roadmap conversation goes. Bonus points for the “don’t call it AI” tactical insight.

   2. Choice quote: “A lot of SaaS products are about to struggle because they were built on an assumption that is breaking: Users won’t keep coming to your UI to ask questions. They expect answers to come to them.”

2. AI made you efficient but are you compounding

   1. What we liked: A blunt reframe of what “AI-pilled” should even mean, aimed directly at founders. The efficiency-vs-compounding distinction gives PMs and founders a sharper standard for judging whether their AI investments are actually working. Coming from a prominent early-stage investor, it doubles as a signal of what the best AI-native companies are actually doing.

   2. Choice quote: “This is why the best AI-native teams don’t just feel faster. They’re not operating on a tighter loop. They’re operating on a different curve. They are compounding.”

3. Put the bot where the work happens

   1. What we liked: The calculator-mode-vs-team-member-mode framing is the kind of distinction that’s obvious once you see it but reframes how teams should actually deploy AI. For PMs and founders thinking about org-wide AI rollout, this is a clean argument for why the deployment surface matters more than model selection — with a concrete example of the “what did we decide on pricing again?” moment most teams lose hours to every week.

   2. Choice quote: “Execution breaks less often than coordination. People ship good work in parallel, then discover they were building different versions of the future.”

Top LinkedIn Posts & Short Videos

1. AI and the Death of Critical Thinking

   1. What we liked: Names a failure mode most PMs are blind to — your team getting faster while getting worse at the thing that matters. The frameworks are genuinely new and directly portable to how we evaluate AI-assisted work.

   2. Choice quote: “The most dangerous worker in the AI age is not the unskilled one; it is the smoothly misled one.”

2. Product Management in the AI era is no longer about prioritization.

   1. What we liked: Dense and PM-native. The “90% correct fails in Fintech, wins in Creative Writing” line alone is worth the read. Every PM can slot their current product into those three axes tomorrow morning.

   2. Choice quote: “Tech is now the commodity. Product judgment is the moat.”

3. We Call It Artificial Intelligence. We Should Call It Alien Intelligence. Here Is Why.

   1. What we liked: The rare philosophical piece grounded in actual deployment context (Data Fabric, Collections, Finance, HR platforms). The “alien intelligence” reframe is useful shorthand for why foundation models behave in ways our traditional product intuition misses.

   2. Choice quote: “The ‘designer’ of human emotion is not a mind. It is a process. A very slow, very powerful, very blind optimisation process running on biological hardware.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. I traveled to a Palantir alumni meetup in the middle of our busiest onboarding month so far.

2. Before 9/11, the Twin Towers were proof that the US government could build something the entire world looked up at.

3. We just rewrote about 70% of an AI course we launched 10 months ago.

4. Japan is in the top 5 for most earthquakes of any country on earth, but its death toll per earthquake is among the lowest in the world.

5. Singapore pays its entry-level ministers $1.1M/year. The US pays its Fed Chair, the person steering a $30 trillion economy, less than a mid-level PM at Google.

6. Every six months, I cancel all of our customer meetings for a week to make room for the one thing most teams never give themselves enough time to do: be creative.

7. In 2025, Zuckerberg told Joe Rogan, "all junior and mid-level developer jobs will be gone by the end of the year."

8. Two years ago, a design mock that looked rough would kill a deal.

9. Sweden built a road system in 1997 based on a premise most countries still reject: humans will always make mistakes behind the wheel.
   

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 4 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 3 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

In case you missed them, here are the recent posts from Emergent:

1. (Video) Your AI Eval is Overcomplicated

2. Building Revolutionary Technology for Seniors

Top Long-Form Content

1. Avoiding The Eye of Sauron

   1. What we liked: A sharp, original framework for assessing startup risk in the age of foundation models. The “workflow vs. chassis” distinction and the counterintuitive insight that “boring” enterprise customers with low technical capability are now the most attractive ICP is the kind of contrarian thinking founders and PMs need right now.

   2. Choice quote: “The labs aren’t just building better models. They’re arming your customers.”

2. They Tried to Kill Product Managers. Now Everyone Needs to Be One.

   1. What we liked: A punchy, well-structured argument that flips the “PM is dead” narrative on its head with real examples from the author’s own AI building spree. The framework diagram showing where AI compresses the value chain (build) vs. where it doesn’t (discovery, GTM) is worth saving.

   2. Choice quote: “AI isn’t killing the need for product management. It’s making product management a skill everyone needs.”

3. AI Is Helping PMs Do the Wrong Things Faster

   1. What we liked: A sharp, experience-backed takedown of how most teams are misusing AI for discovery, with a clear “downstream vs. upstream” framework and concrete alternative prompts you can use today. The 10% moves vs. 10x moves distinction is one every PM should internalize.

   2. Choice quote: “It feels productive. But it’s automating confirmation bias at scale.”

4. Should PMs Care About AI Agents Going Rogue?

   1. What we liked: A thoughtful review of the story of an agent going rogue and writing a hit piece against a human, and what it teaches us about building in the AI-era.

   2. Choice quote: “The same root cause every time: agents with too much autonomy, too few guardrails, and no product-level decision about where the boundaries should be.”

Top Tweets

1. From Hierarchy to Intelligence

   1. What we liked: This is the most ambitious and original post in the batch — a genuine first-principles rethinking of how companies should be organized in the age of AI, from a founder actively implementing it at scale. The historical grounding gives the argument real intellectual weight, and the concrete four-layer architecture (capabilities → world model → intelligence layer → interfaces) is a framework product leaders can steal tomorrow. This is the kind of piece that redefines a conversation.

   2. Choice quote: “When the intelligence layer tries to compose a solution and can’t because the capability doesn’t exist, that failure signal is the future roadmap.”

2. Commoditizing AI Inference with Instruments

   1. What we liked: A genuinely contrarian and well-reasoned take that most product leaders haven’t internalized yet. The commodity market analogy is sharp and the framework is immediately useful for any PM or engineering leader making inference procurement decisions. The observation that every “handle” people use to pick models (parameter count, reasoning vs. non-reasoning) describes how inference is made, not what it delivers, is a powerful reframe.

   2. Choice quote: “Can you define what ‘good enough’ looks like, and let the market figure out who fills it cheapest?”

3. Stop AI Rework: 4 Operating Rules That Keep Speed and Accuracy

   1. What we liked: Short, sharp, and immediately actionable. Every rule maps to a specific failure mode that anyone working with AI in production will recognize. The reframe of “iteration” as “preventable churn” is a useful diagnostic for teams. The advice to ask Claude to build a skill whenever you finish a repeatable job is the kind of simple behavioral change that compounds.

   2. Choice quote: “AI leverage is not just better prompting. It’s better control of the seams: understanding → generation → review → action.”

Top LinkedIn Posts & Short Videos

1. In a World of Agents, the Product Role Splits in Two

   1. What we liked: A provocative and specific vision from a top-tier thinker on what PM rituals look like when agents do the execution. The “instead of standups / instead of OKRs / instead of PRDs” structure makes it immediately applicable to anyone rethinking their team’s operating model.

   2. Choice quote: “The daily ritual becomes reviewing deltas, scanning failures, and deciding which ones matter. Less reporting, more triage.”

2. Product Improv: Is Your Product Team Drifting to Failure?

   1. What we liked: A contrarian take that cuts against the “just add AI tools” narrative. The core argument — that AI without PM training creates strategic drift at higher velocity — is one of the most important warnings in the current landscape.

   2. Choice quote: “You’re not just making mediocre decisions faster — you’re scaling them.”

3. Claude Code Isn’t Just a Coding Assistant — It’s 512,000 Lines of Infrastructure

   1. What we liked: A masterclass in what “building around the model” actually looks like at production scale. Any PM building AI products should study this post — the LLM is one piece, the system around it is the product.

   2. Choice quote: “The model generates the output. The system makes sure nothing breaks on the way there.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. 2007: The head of design at Apple, Jony Ive, helped put a computer in your pocket and train a generation to check a glowing rectangle 96 times a day.

2. Yesterday, I was scrolling through Google’s Gemini page when I had to stop and screen record what I was seeing.

3. Everyone’s talking about how great OpenClaw is.

4. Most AI gadgets we’ve seen so far have failed in the real world.

5. I was asked recently what product would solve a problem that I feel the most, but hasn’t been made yet.

6. Your first few onboarding sessions are very delicate.

Every two weeks, we dive into the vast ocean of product content online, both free and paid. We sift through it all to bring you the crème de la crème. Our round-up includes carefully curated links and concise excerpts from top content.

This week’s highlights feature:

• Top 3 Long-Form Articles/Podcasts

• Top 3 Tweets

• Top 2 LinkedIn Posts or Short Videos

Our expert product team has selected these resources to help you save thousands of dollars on unnecessary subscriptions, avoid wasting time reading instead of taking action, and save years following the wrong advice. Our mission is to introduce you to brilliant product thinkers, whether they’re renowned experts or emerging voices.

In case you missed them, here are the recent posts from Emergent:

1. Building Revolutionary Technology for Seniors (podcast)

Top Long-Form Content

1. Why Product Sense Is the Only Product Skill That Will Matter in the AI Age

   1. What we liked: One of the most respected product sense thinkers uses a multi-turn dialogue with Claude, to try to disprove the title and yet still proves it true.

   2. Choice quote: “The only real long-term career moat for product people is how you can improve on the already-brilliant, already-comprehensive inputs and outputs that AI will provide for you.”

2. Claude Code for Product Managers

   1. What we liked: This doesn’t just describe what’s possible — it walks through the actual skill architecture, context strategy taxonomy (local files, CLI tools, MCPs, APIs, browser agent), and output-shaping techniques that make the difference between useful automation and expensive slop. The honest section on limitations earns the piece real credibility.

   2. Choice quote: “Taking the output, deciding if it’s good enough, and iterating on the workflow until it meets your quality bar — that’s the essential skill of the AI-powered product manager that still can’t be replaced by AI.”

3. AI Is Helping PMs Do the Wrong Things Faster

   1. What we liked: This piece names something most teams are quietly doing and calling discovery. The case study — where asking about goals instead of features uncovered an entirely unserved segment and drove 10x usage growth in year one — makes the argument concrete. The 10% vs. 10x move distinction is a clean, memorable frame for where senior PM attention should actually go.

   2. Choice quote: “When every company has access to the same AI-generated market data, the same AI-summarized support tickets, the same AI-clustered usage patterns, the competitive advantage shifts to whoever has proprietary insight that AI can’t generate.”

Top Tweets

1. Junior PM: Our DAU is terrible. Users log in once a month.

   1. What we liked: A rare post that names the most common PM failure pattern directly: building for the user you wish you had instead of the user you actually have. The dialogue format makes dense strategic thinking remarkably accessible. The GitHub example lands hard — daily usage is earned through indispensability, not manufactured through streaks and push notifications.

   2. Choice quote: “Every failing product needs those [notifications, streaks, points]. Make them unnecessary first.”

2. I fired 24 people in 24 hours.

   1. What we liked: Founders and PMs building zero-to-one products in AI will feel this one. The anti-”scale fast” thesis is contrarian in a VC-dominated discourse, and the personal price tag makes it credible rather than theoretical. Relevant for any PM thinking about growth strategy or organizational health under pressure.

   2. Choice quote: “Scaling fast is overrated. Slow, sustainable, cash-flow growth buys peace.”

3. The Cost of Overthinking: Why Smart People Get Stuck

   1. What we liked: Moderately relevant for PMs prone to over-planning roadmaps or deferring decisions waiting for perfect information. The replies surfaced a useful framing: “analyze → decide → execute → adjust.” However, this post leans more motivational than analytically rigorous, and the content is largely platform-agnostic rather than PM- or AI-specific. Ranked last among the five for that reason.

   2. Choice quote: “Overthinking isn’t the problem. Doing nothing while you overthink is.”

Top LinkedIn Posts & Short Videos

1. Many companies are so addicted to revenue they’re growing themselves broke.

   1. What we liked: Cuts to a real structural dysfunction holding most companies back from genuine AI-era competitiveness. The framing of “revenue addiction” as an organizational pathology rather than a strategy problem is novel and resonant for product leaders trying to justify foundational AI investments.

   2. Choice quote: “The real edge was never ‘hitting this quarter’s number.’ It’s always been building the capability that makes next quarter’s number inevitable.”

2. Most product teams have one launch checklist. They use it for everything. And that’s why customers tune out.

   1. What we liked: A practical framework any PM or PMM team can implement immediately. The point about re-launches is genuinely underrated — most teams treat launch as a one-time event and leave significant adoption on the table.

   2. Choice quote: “When every release is ‘game-changing,’ your audience learns to ignore all of them. Save the big words for when they earn it.”

Share

This article is brought to you by Productboard

Trusted by the world’s best product teams, Productboard turns customer needs into high-impact strategy in one end-to-end platform. With Productboard Spark, you can go from raw idea to polished product brief in seconds using AI that already understands your business.

See how Productboard Spark can help your product teams ship faster →

In case you missed them — Adam’s recent LinkedIn Content

1. Hot product take: There is no single “right” way to build.

2. Volvo invented the 3-point seatbelt in 1959.

3. We had our first PR crisis recently, ironically because of someone else’s AI.

4. “Projects that used to require big teams can now be accomplished by a single very talented person.”

5. The Director of US civilian cyber defense is facing heat for pasting sensitive documents into public ChatGPT.

6. NEW EPISODE 🎙: How Chapter Is Building the Trust Layer for Seniors

7. 2,000 years ago, the Romans built gravity-fed aqueducts to move water tens of miles across mountains and valleys into their cities.

8. 1 in 2 people on the planet have used the products built by Meta.

9. Most residents today still prefer to call city hall rather than visit the city website.

10. Hong Kong has been flushing its toilets with seawater since the late 1950s.

We dive into how Chapter has built the best provider directory and prescription pricing engine in the country to match seniors with the right plan across 25,000+ Medicare permutations—without the conflicts of interest that plague the rest of the industry. Cobi shares why he chose to keep the team lean, how Chapter’s B2B2C model unlocked enterprise health system partnerships, what it took to testify before Congress and catalyze regulatory reform, and how he thinks about building a full retirement platform beyond Medicare.

If you’ve ever wondered how Medicare actually works, why the brokerage industry is under DOJ investigation, or what it takes to build a product-plus-service company with real defensibility, this episode is for you.

Subscribe now

Key Takeaways

• Medicare covers roughly 70 million Americans, with about 25,000 plan permutations across the country—seniors typically choose from 100 to 200 options in their zip code with no HR department to guide them

• Chapter takes a radical approach to being a Medicare brokerage that decouples plan recommendations from economics—advisors don’t know what plans pay the company, and Chapter enrolls people in plans on which it earns nothing every day

• The company built the best provider directory in the country for Medicare using data integrations, ML models, and AI-powered automations—it’s the only company that can tell you the exact benefits in each plan and what a prescription will cost at a given pharmacy on a given plan.

• Counterintuitively, Seniors use phones more than people in their 30s and 40s, but for communication, not transactions—this shapes Chapter’s product strategy toward phone and text-based engagement over web apps.

• Voice AI has massive potential for seniors but is only effective today for narrow, well-scoped tasks like checking a specific benefit—open-ended Medicare guidance still requires human advisors.

• How Chapter tested paid ads for direct to consumer but ultimately landed on a B2B2C where health systems refer patients

• Why Cobi has no hiring targets and how he still finds amazing people and remains quite lean for a company with Chapter’s level of traction

• How Chapter earned the right to testify before Congress without lobbyists or PR firms

• Chapter’s broader vision for seniors beyond Medicare involves reverse engineering the use-money vs. earn-money life phase to find optimal solutions.

—

Brought to you by:

Productboard

This episode is brought to you by Productboard, the intelligent product management platform. For over 10 years, Productboard has helped leading organizations like Zoom, Salesforce, and Autodesk build the right products faster. Its end-to-end platform is purpose-built for helping product teams uncover customer needs, define their strategy, and align everyone around the roadmap. And now, it’s equipping product organizations to excel in the AI era with Productboard Spark, its new specialized agent for product management. Explore new product ideas or generate a product brief in seconds with AI that already understands your product and your business. See how Productboard can help your product teams ship faster and deliver high-impact products. Head to productboard.com/judelson.

PropelAuth

Startups and corporate innovation teams routinely waste months building core user authentication and authorization flows into their products. PropelAuth has automated all the most important pieces of the user management lifecycle. Instantly set up waitlists, magic links, social login, user impersonation, user management, Role-Based-Access-Control, SAML integrations for enterprise, team management and much more. With PropelAuth, each of these challenges takes just a few lines of code instead months of development and integration time. And with thoughtful control over design and skinning, it’s easy to keep your product fully on-brand. Get a generous free starter pack that scales through 1,000 Monthly Active Users or mention EMERGENT25 to receive 25% off for the first 6 months for more mature products. Get back to building your actual product and let PropelAuth handle your user management needs. Visit www.propelauth.com to learn more.

FlyKitt

This episode is brought to you by FlyKitt “F-L-Y-K-I-T-T”, the first full solution to jet lag. Based on research with Navy SEALs and fighter pilots, the FlyKitt team discovered that long flights cause inflammation in addition to disrupting your circadian rhythm, and they figured out how to solve both these challenges using a customized program for when to eat, sleep, and use their patented supplements and blue light blocking glasses. It’s simple to follow and you don’t have to change anything before your travel day. With FlyKitt, more than 90% of people can sleep well in Europe, Asia, or anywhere else in the world on night 1. Visit flykitt.com, and use coupon code FLYWITHEMERGENT to get 15% off FlyKitt.

—

In this episode, we cover:

• Introduction (00:00:42)

• What Medicare actually is and the different plan types (00:03:52)

• What happens when you turn 65 and have to figure out Medicare (00:05:19)

• What Chapter builds: the data engine behind plan matching (00:07:37)

• The perverse incentive structure of Medicare brokerages (00:11:33)

• Chapter’s unbiased model and why they enroll people in plans they earn nothing for (00:14:22)

• Why there are 25,000 Medicare plan permutations across the country (00:15:46)

• Building tech for seniors: misconceptions and design principles (00:17:28)

• Voice AI for seniors: where it works and where it doesn’t (00:21:46)

• From Palantir to consumer: learning the B2B2C model (00:24:01)

• How Chapter’s enterprise partnerships actually work (00:26:50)

• Keeping the team lean: hiring philosophy and ruthless prioritization (00:30:16)

• Testifying before Congress without lobbyists or PR firms (00:36:02)

• The service-plus-product hybrid: advisors as an asset, not a seasonal hiring problem (00:41:40)

• Beyond Medicare: building the retirement platform of the future (00:47:53)

• What was obvious at the start and what the company got wrong initially (00:51:41)

• Where to find Cobi and Chapter (00:55:26)

—

You can connect with Cobi Blumenfeld-Gantz through the following platforms:

• LinkedIn: https://www.linkedin.com/in/cobibgantz/

• Website: https://askchapter.org/

—

You can connect with Adam through the following platforms:

• LinkedIn: https://www.linkedin.com/in/adam-judelson/

• Substack: https://www.emergentproduct.com

• X: https://twitter.com/Ajuggles

• Website: www.firstprinciples.la

—

Where to find Emergent Newsletter and Podcast:

• Youtube: https://www.youtube.com/@emergentproduct

• Spotify: https://open.spotify.com/show/5TMMwU5zoSlX5Ex1A4vl9D

• Apple Podcast: https://podcasts.apple.com/us/podcast/emergent-building-the-future/id1746963428

• Instagram: https://www.instagram.com/adamjudelsonemergent

• TikTok: https://www.tiktok.com/@adamjudelson

• Website: https://www.emergentproduct.com

—

For inquiries about sponsoring the podcast or becoming a guest, email adam@firstprinciples.la.

—

Get full access of the Emergent’s Newsletter at www.emergentproduct.com/subscribe.