# Almost Timely News - The Biggest Problem with AI Today > [!question] Lingering Questions > - The 29%-org-ROI-vs-5x-individual-productivity juxtaposition compares different units (org-level vs. self-reported individual) and is presented rhetorically rather than arithmetically — worth flagging whenever this stat gets reused. > - The generalization from Reddit verbatims (180 days, 40+ subreddits) to "most people" is inductive and unquantified — a real pattern, but not a measured one. > - The "executive function defines our sentience" framing leans on an appeal-to-nature move; worth separating the sentience claim from the actual PODS (Plan/Organize/Decide/Solve) framework, which stands without it. > [!tip] Article Ideas > - "The two-question gate" as a standalone editorial-review checklist item: does AI make the output better, AND does it make me better? Both must be yes. > - Cross-post pairing this with [[3. Resources/Literature Notes/The Problem with Streaming and the Case for Physical Media]] (Rizzuto) — media ownership and cognitive ownership as the same argument in two domains. > [!link] Cross Reference > - [[3. Resources/Literature Notes/The Problem with Streaming and the Case for Physical Media]] — same publish window (2026-07-05), parallel "don't outsource what makes it yours" argument > - [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] — executive-function abdication as a knowledge-work risk, not just an individual-habit problem > [!quote] Vocabulary > - **PODS** — Penn's operational handle for executive function: Plan, Organize, Decide, Solve > - **Metacognition** — "thinking about thinking"; Penn's reframing of critical thinking as the actual skill at stake in AI delegation ## Bibliographic capsule Christopher S. Penn argues that AI's biggest problem is neither cost, sustainability, nor ethics — those are symptoms — but abdicated executive function: after loading 180 days of 40+ marketing/AI subreddits (~800,000 words) into NotebookLM and cross-referencing against his own newsletters via Claude Code, he reframes ten distinct Reddit-sourced AI pain points as a single root failure: "we've stopped thinking about thinking." ## Author's thesis Every symptom commonly blamed for AI's problems — invented visibility metrics, degrading agentic oversight, ROI claims that don't reconcile, wasted budget on default models, rental-not-ownership SaaS dependence, sycophancy, broken AI detectors, task-by-task organizational hollowing, tokenmaxxing, unpaid marketer-as-trainer labor — is actually one failure repeated ten times: nobody is exercising Plan/Organize/Decide/Solve judgment, so nobody is accountable for the outcome. ## Key arguments (≤3) 1. **The thesis is performed, not just stated.** When Claude's own analysis nudged Penn toward a "measurement is broken" framing, he explicitly overrode it — noting measurement is a trailing indicator and a symptom of the deeper problem — making the override itself an act of the executive function the article argues for. 2. **Capability and danger coexist without contradiction.** Penn concedes well-prompted AI can out-plan, out-organize, out-decide, and out-solve humans at most language-based tasks — and argues the real danger isn't that AI is bad at this, but that ceding the function wholesale removes the accountable human. 3. **The antidote is procedural, not abstinence.** For tasks that matter, lead first (dictate your own analysis, draft your own take) and force the machine to educate you second — rather than accepting machine output as a starting point. ## Claim atoms ### AI's real problem is abdicated executive function > [!quote] ID:rw_penn20260705a > We've abdicated far too much of our executive function to AI. We've surrendered our thinking. Where/why: the article's opening thesis statement. Reframes cost, sustainability, and ethics as symptoms of a cognition-and-accountability failure — the article's spine. ### Lead first, then force the machine to educate you > [!quote] ID:rw_penn20260705b > For every task that matters, always start with something you lead, and force the machines to educate you. Where/why: "The Antidote" section. The prescriptive core — use AI to grow expertise, not substitute for it. ### Two-question gate: better output and better you > [!quote] ID:rw_penn20260705c > 1. Does the use of AI make the output better? 2. Does the use of AI make me better? Where/why: the closing section. A portable decision test — both must be yes, or the tool is being used badly. ## Permanent candidates - **[P1] Outsourcing judgment removes the accountable decision-maker** — supported by H1; checked existing Permanents, found related-but-distinct claims on abdication and empowerment ([[3. Resources/Permanent Notes/Empowerment Without Clarity Is Abdication]]); no exact duplicate, promoted below. ## Processing checklist - [x] Confirmed this doc's `notes` field already contained a full deep-read (Refactor Appendix present) — ported directly - [x] Checked for an existing Permanent note before promoting P1 — found none exact, created with cross-link to the related note - [x] Connect into the relevant MOC — [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] ## Links - [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] — executive-function abdication as the article's central knowledge-work risk - [[3. Resources/Literature Notes/The Problem with Streaming and the Case for Physical Media]] — companion source, same publish window, parallel argument in a different domain - [[3. Resources/Readwise/Articles/Almost Timely News 🗞️ the Biggest Problem With AI Today]] — source document