Two new cards, two new operators, no new patterns. Three runs across June 26-28 captured the bulk of recent output from the scan roster, leaving a slim window for today. What cleared the bar: two independent claims about decision quality under uncertainty, arriving from different disciplines.
Theme 1, The scarce resource shifts from knowledge to clarity
Joe Hudson published "The new inner game: Your unfair advantage in the age of AI" in Lenny's Newsletter on June 23 (Emotional clarity becomes the scarce competitive advantage as AI commoditizes knowledge and effort). Hudson coaches the research team at OpenAI and works with founders across Apple, Google, and X. His argument is direct: knowledge and effort used to be the primary competitive advantages. AI is making both nearly free. The capacity that remains scarce is emotional clarity, the ability to feel without being controlled by what you feel.
The mechanism matters. When leaders avoid difficult emotions, their decision space contracts. A CEO who cannot face disappointing someone becomes a CEO who cannot kill failing projects. The roadmap clogs. The avoidance shows up as reasonableness, not as fear. Hudson's coaching addresses the internal constraint, not the surface behavior.
"When knowledge and effort are nearly free, emotional clarity is scarce. But luckily, it's a skill that can be learned."
He grounds this in neuroscience: choices are fundamentally emotional, with rational processing acting downstream of emotional context. Improving that layer raises the quality of every decision above it. The argument is not that AI is bad or that effort no longer matters. It is that the comparison class has changed. "You can't out-grind a server farm," Hudson writes. So effort-based identity is now a liability. What scales instead is clarity about what to build, what to stop, and who to trust.
- Joe Hudson, Emotional clarity becomes the scarce competitive advantage as AI commoditizes knowledge and effort. As AI compresses the cost of knowledge and effort, emotional clarity becomes the scarce and learnable differentiator for leaders and knowledge workers.
Theme 2, Confidence scores in prioritization frameworks measure nothing real
Jason Cohen published "Lost confidence" at A Smart Bear on June 21 (RICE-style confidence scores are mathematically undefined and empirically uncorrelated with outcomes). Cohen founded WP Engine and scaled it past $100M ARR. His target is RICE and similar frameworks where a "confidence" multiplier adjusts a project's priority score. His argument: confidence cannot be defined, cannot be measured, and shows no empirical correlation with outcomes. A framework built on it is not noisy; it is arbitrary.
Three failure modes stack. The definitional problem: a "30% confidence" score requires calibration across historical predictions to carry meaning. No team maintains this. The empirical problem: across many product cycles, Cohen observes no correlation between stated confidence at kickoff and actual outcomes at delivery. The structural problem: high-impact, high-uncertainty bets score low on confidence by construction, so the framework deprioritizes exactly the projects worth making asymmetric bets on.
"Stop pretending you can quantify confidence, or even define it. Instead, use techniques that work whenever the future is unpredictable. Because it always is."
His alternatives are designed for genuine uncertainty: always-true features (valued by most of the customer base regardless of market conditions), asymmetric bets (bounded downside with large potential upside), and quick discovery through small complete prototypes. Each technique works when the future is opaque, not by pretending it is knowable.
The two cards are not in conflict. Hudson addresses the internal quality of judgment. Cohen addresses the framework used to act on it. Both are arguing that false precision is more dangerous than acknowledged uncertainty.
- Jason Cohen, RICE-style confidence scores are mathematically undefined and empirically uncorrelated with outcomes. RICE-style confidence scores are undefined inputs: they cannot be calibrated, show no empirical correlation with outcomes, and structurally bias portfolios against high-upside uncertain bets.