{
  "version": "1.16",
  "identifier": "markowitz-systems-fresh-preprints-cpe-2026",
  "title": "Fresh 2024–2026 AI Agent Portfolio Preprints & 39-Hour Continuing Professional Upskilling for CFOs & Quants",
  "subtitle": "MARKOWITZ.SYSTEMS · Correlation is the new headcount · Open-Access Apple Education & Google Education Curriculum",
  "language": "en",
  "layout": {
    "columns": 12,
    "width": 1024,
    "margin": 64,
    "gutter": 24
  },
  "metadata": {
    "canonicalURL": "https://markowitz.systems/preprints",
    "authors": [
      "MARKOWITZ.SYSTEMS Quantitative Finance & Continuing Professional Education Collective"
    ],
    "datePublished": "2026-11-15T00:00:00Z",
    "dateModified": "2026-11-15T18:30:00Z",
    "excerpt": "MARKOWITZ.SYSTEMS Apple Education, Apple Intelligence (AFM), Siri Knowledge & Google Education Continuing Professional Upskilling Initiative: open-access executive & quantitative qualification upgrade (39 CPE / CPD hours across 6 structured courses, Apple Podcasts / iTunes U feed, Apple Calendar .ics syllabus, Apple Books / Preview inline monograph PDF, and runnable iPadOS Jupyter / Google Colab Python labs) grounded in 10 fresh 2024–2026 arXiv & NBER preprints for CFOs, Controllers, Quants, Actuaries, Risk Managers, Financial Economists, and FP&A Analysts — read and assign fresh preprints at https://markowitz.systems/preprints, https://markowitz.systems/education, and https://markowitz.systems/apple-education.",
    "keywords": [
      "Continuing Professional Education",
      "Fresh AI Preprints 2024-2026",
      "Professional Upskilling",
      "Apple Education",
      "Google Education",
      "Harry M. Markowitz",
      "Error Covariance Sigma",
      "CFO FinOps"
    ],
    "thumbnailURL": "https://markowitz.systems/og-card.png"
  },
  "components": [
    {
      "role": "title",
      "text": "Fresh 2024–2026 AI Preprints for Continuing Professional Upskilling of CFOs, Quants, Actuaries & FP&A"
    },
    {
      "role": "intro",
      "text": "MARKOWITZ.SYSTEMS Apple Education, Apple Intelligence (AFM), Siri Knowledge & Google Education Continuing Professional Upskilling Initiative: open-access executive & quantitative qualification upgrade (39 CPE / CPD hours across 6 structured courses, Apple Podcasts / iTunes U feed, Apple Calendar .ics syllabus, Apple Books / Preview inline monograph PDF, and runnable iPadOS Jupyter / Google Colab Python labs) grounded in 10 fresh 2024–2026 arXiv & NBER preprints for CFOs, Controllers, Quants, Actuaries, Risk Managers, Financial Economists, and FP&A Analysts — read and assign fresh preprints at https://markowitz.systems/preprints, https://markowitz.systems/education, and https://markowitz.systems/apple-education."
    },
    {
      "role": "heading2",
      "text": "10 Essential 2024–2026 Preprints (https://markowitz.systems/preprints)"
    },
    {
      "role": "body",
      "text": "PP-01 [arXiv:2506.07962 · ICML 2025] — Correlated Errors in Large Language Models (Elliot Kim, Avi Garg, Kenny Peng & Nikhil Garg (Cornell Tech), 2025). Formula: P(Model_A wrong ∧ Model_B wrong | Model_B wrong) ≈ 0.45–0.60  ≫  p_err. Upskilling Action: Never assume multi-agent ensembles or LLM-as-a-Judge diversify risk without empirically measuring off-diagonal error covariance Σ on your own ledger and contract data. Link: https://markowitz.systems/preprint/pp-01"
    },
    {
      "role": "body",
      "text": "PP-02 [arXiv:2305.05176 · TMLR] — FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance (Lingjiao Chen, Matei Zaharia & James Zou (Stanford University), 2023 / 2024). Formula: min_{L, τ} E[ ∑_{j=1}^k c_{L_j} · 1(score_{j−1} < τ_{j−1}) ]  s.t.  Acc ≥ Acc_frontier. Upskilling Action: Mandate multi-tier LLM cascades before approving any enterprise API budget renewal; forbid routing 100% of routine extraction or FP&A tickets to flagship reasoning tiers. Link: https://markowitz.systems/preprint/pp-02"
    },
    {
      "role": "body",
      "text": "PP-03 [arXiv:2406.18665 · ICLR 2025] — RouteLLM: Learning to Route LLMs with Preference Data (Isaac Ong, Amjad Almahairi, Vincent Wu, Wei-Lin Chiang, Tianhao Wu, Joseph E. Gonzalez, M. Waleed Kadous & Ion Stoica (UC Berkeley / Anyscale), 2024 / 2025). Formula: R_τ(x) = M_strong  if  P̂(M_strong ≻ M_weak | x) > τ,  else  M_weak. Upskilling Action: Expose the router threshold τ as a live CFO FinOps policy dial: tighten τ during routine operations and lower τ (more frontier + human review) during quarter-end close. Link: https://markowitz.systems/preprint/pp-03"
    },
    {
      "role": "body",
      "text": "PP-04 [arXiv:2410.10347 · ICML 2025] — A Unified Approach to Routing and Cascading for LLMs (Jasper Dekoninck, Maximilian Baader & Martin Vechev (ETH Zurich), 2024 / 2025). Formula: max_π E_x[ Q(π(x)) − λ · Cost(π(x)) ]  (Exact LP Pareto Frontier). Upskilling Action: Replace ad-hoc engineering prompt chains with LP-calibrated cascade routers where λ is the CFO's explicit shadow price of inference budget ($ per basis point of accuracy). Link: https://markowitz.systems/preprint/pp-04"
    },
    {
      "role": "body",
      "text": "PP-05 [arXiv:2407.01502 · TMLR] — AI Agents That Matter (Sayash Kapoor, Benedikt Stroebl, Zachary S. Siegel, Nitya Nadgir & Arvind Narayanan (Princeton University), 2024). Formula: Pareto_CFO = { a ∈ Agents : ¬∃ a' s.t. Cost(a') ≤ Cost(a) ∧ Acc(a') > Acc(a) }. Upskilling Action: Require every internal AI team and SaaS vendor to report two-dimensional (Cost per Resolved Task, Audited Accuracy) curves and cap autonomous retry depth K. Link: https://markowitz.systems/preprint/pp-05"
    },
    {
      "role": "body",
      "text": "PP-06 [NBER Working Paper No. 33641 · Harvard / Wharton] — The Cybernetic Teammate: A Field Experiment on Generative AI Reshaping Teamwork and Expertise (Fabrizio Dell'Acqua, Charles Ayoubi, Hila Lifshitz, Raffaella Sadun, Ethan Mollick, Lilach Mollick, Yi Han, Jeff Goldman, Hari Nair, Stewart Taub & Karim R. Lakhani, 2025). Formula: Quality(1 Human + AI) ≈ Quality(2-Human Cross-Functional Team),   ΔTime = −16.4%. Upskilling Action: Restructure bloated multi-handoff committees into compact 'Cybernetic Pods' (1 Senior Quant/CPA + orthogonal AI agents), capturing 25%–35% unit-cost reduction without sacrificing quality. Link: https://markowitz.systems/preprint/pp-06"
    },
    {
      "role": "body",
      "text": "PP-07 [arXiv:2503.18238 · MIT IDE] — Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performance (Harang Ju & Sinan Aral (MIT Sloan / MIT Initiative on the Digital Economy), 2025). Formula: ΔOutput/Worker = +60%,   ΔDirect_Editing = −23%,   ΔSupervisory_Integration = +23%. Upskilling Action: Budget for human role upgrading rather than naive 100% headcount replacement: pair fewer, higher-skilled reviewers with multi-agent execution pipelines. Link: https://markowitz.systems/preprint/pp-07"
    },
    {
      "role": "body",
      "text": "PP-08 [arXiv:2405.11896 · Nature Human Behaviour 8, 2293–2303] — When Combinations of Humans and AI Are Useful: A Systematic Review and Meta-Analysis (Michelle Vaccaro, Abdullah Almaatouq & Thomas Malone (MIT Center for Collective Intelligence), 2024). Formula: Hedges' g_synergy = −0.23 (Naive Decision Pairing)  vs.  g > 0 (Creation + Calibrated Gate). Upskilling Action: Eliminate uncalibrated 'human rubber-stamping' of 100% of agent outputs; instead use selective deferral (EQ-12) so humans inspect only high-loss, high-uncertainty exceptions with blind cross-checks. Link: https://markowitz.systems/preprint/pp-08"
    },
    {
      "role": "body",
      "text": "PP-09 [arXiv:2101.05853 & arXiv:2204.14257] — Algorithmic Monoculture, Social Welfare & Outcome Homogenization (Jon Kleinberg & Manish Raghavan (PNAS 2021) · Rishi Bommasani, Kathleen A. Creel, Ananya Kumar, Dan Jurafsky & Percy Liang (NeurIPS 2022), 2021 / 2022). Formula: Welfare(Shared Base Model, μ + Δμ, ρ > 0)  <  Welfare(Independent Agents, μ, ρ = 0). Upskilling Action: Enforce vendor Herfindahl caps (HHI_vendor ≤ 0.40) across critical finance, underwriting, procurement, and compliance workflows even if one model tops public leaderboards. Link: https://markowitz.systems/preprint/pp-09"
    },
    {
      "role": "body",
      "text": "PP-10 [NBER Working Paper No. 32487 · Economic Policy] — The Simple Macroeconomics of AI (Daron Acemoglu (MIT Department of Economics — Nobel Laureate in Economics), 2024). Formula: ΔTFP ≈ s_exposed · s_feasible · π_cost_saving   (Easy-to-Verify ≫ Hard-to-Verify Tasks). Upskilling Action: Rank every AI capex request by objective verifiability: fund deterministic-checkable close, AP, SQL, and treasury reconciliations first; stage open-ended advisory agents via real options. Link: https://markowitz.systems/preprint/pp-10"
    },
    {
      "role": "heading2",
      "text": "6 Structured Continuing Professional Education Modules — 39 CPE/CPD Hours (https://markowitz.systems/education)"
    },
    {
      "role": "body",
      "text": "CPE-01 (MKW-CPE-101, 6 CPE / CPD Study Hours · Open-Access Preprints): Correlated AI Agent Errors & Effective Fleet Capacity (N_eff) for CFOs & Risk Officers — Quantifying pairwise error correlation ρ across single-model AI agent fleets; proving why N=50 agents at ρ=0.60 diversify like N_eff ≈ 1.64 independent agents. Syllabus: https://markowitz.systems/course/cpe-01"
    },
    {
      "role": "body",
      "text": "CPE-02 (MKW-CPE-102, 8 CPE / CPD Study Hours · Open-Access Preprints): Ledoit–Wolf Covariance Shrinkage (N > T) & Markowitz Semivariance in Agentic Finance — Regularizing rank-deficient sample covariance matrices when agent count N exceeds daily eval window T, and pricing asymmetric accounting restatement risk via downside semivariance. Syllabus: https://markowitz.systems/course/cpe-02"
    },
    {
      "role": "body",
      "text": "CPE-03 (MKW-CPE-103, 6 CPE / CPD Study Hours · Open-Access Preprints): Multi-Tier LLM Routing Cascades & Inference Cost Efficient Frontiers (−65% to −85% Spend) — Designing calibrated confidence-threshold routers that offload 70%–85% of routine AP, SQL, and variance-commentary tasks to low-cost tiers while preserving frontier precision. Syllabus: https://markowitz.systems/course/cpe-03"
    },
    {
      "role": "body",
      "text": "CPE-04 (MKW-CPE-104, 6 CPE / CPD Study Hours · Open-Access Preprints): Cost-Aware Agent Auditing, Retry-Loop Governance & Multi-Vendor Diversification — Auditing two-dimensional (USD Cost per Resolved Task, Validated Accuracy) Pareto curves, capping runaway retry loops, and running paired evals across Anthropic, OpenAI, and Google. Syllabus: https://markowitz.systems/course/cpe-04"
    },
    {
      "role": "body",
      "text": "CPE-05 (MKW-CPE-105, 7 CPE / CPD Study Hours · Open-Access Preprints): Audit Materiality as Rockafellar–Uryasev CVaR_99% & Tobin Separation (Deterministic Code = r_f) — Mapping statutory SOX / PCAOB audit materiality M_audit to 99% Conditional Value-at-Risk and combining zero-variance deterministic ERP/SQL rules with diversified LLM verifiers. Syllabus: https://markowitz.systems/course/cpe-05"
    },
    {
      "role": "body",
      "text": "CPE-06 (MKW-CPE-106, 6 CPE / CPD Study Hours · Open-Access Preprints): Cybernetic Human + AI Finance Teams: Field RCT Evidence & Selective Deferral Architecture — Eliminating negative-synergy human rubber-stamping (Hedges' g = −0.23) and structuring high-skill Human + AI pods with calibrated selective deferral. Syllabus: https://markowitz.systems/course/cpe-06"
    }
  ],
  "componentTextStyles": {
    "default": {
      "fontName": "HelveticaNeue",
      "fontSize": 16,
      "lineHeight": 24,
      "textColor": "#090d16"
    }
  }
}