Rosa Del Mar

Daily Brief

Issue 79 2026-03-20

Mobilization As Pre-Commitment And Whole-Of-Economy Capacity

Issue 79 Edition 2026-03-20 9 min read
General
Sources: 1 • Confidence: Medium • Updated: 2026-03-25 17:51

Key takeaways

  • Sankar claims Israel's post-October 7 mobilization brought back reservists with decades of industry experience who drove rapid technology modernization in the IDF within months.
  • Sankar asserts that the economic and societal impact of AI will be determined by human choices in how it is applied, making outcomes depend on wielders more than inventors.
  • Sankar asserts that AI and modern tooling enable domain experts in uniform to build operational software quickly, shifting innovation from PowerPoint proposals to empirical prototypes.
  • Sankar argues that the primary harm of post-Cold War defense consolidation was conformity and financialization more than reduced competition among vendors.
  • Sankar asserts that AI will pressure 'beta' software that makes organizations more similar, while 'alpha' software that expresses competitive differentiation will gain value.

Sections

Mobilization As Pre-Commitment And Whole-Of-Economy Capacity

  • Sankar claims Israel's post-October 7 mobilization brought back reservists with decades of industry experience who drove rapid technology modernization in the IDF within months.
  • Sankar asserts that effective wartime mobilization requires the whole country's industrial and technological base, not only the specialized defense industrial base.
  • Sankar asserts that commercial mass-market production can subsidize national security capability by pulling defense-relevant technology down a price-performance curve via private-sector R&D scale.
  • Sankar asserts that innovation depends on domestic production because producing tightens R&D-to-factory feedback loops and accelerates iteration.
  • Sankar asserts that World War II mobilization advantage depended on pre-war preparation mechanisms (e.g., Lend-Lease and factory retooling lead time) rather than a purely post-Pearl Harbor surge.
  • Sankar suggests voluntary U.S. civil-military fusion could be expanded by reactivating WWII-era direct-commission authorities to bring technical professionals into the military at scale.

Agency-First Ai Governance And National Competitiveness Constraints

  • Sankar asserts that the economic and societal impact of AI will be determined by human choices in how it is applied, making outcomes depend on wielders more than inventors.
  • Sankar argues AI inventors' doomer narratives should not be treated as definitive because model-building skill does not imply correctness about societal implications and human agency can steer outcomes.
  • Sankar argues the biggest U.S. risk in the AI race is self-inflicted failure driven by loss of national will, focus, and institutional legitimacy rather than external attack.
  • Sankar asserts reindustrialization will be asymmetric because AI-enabled manufacturing can make workers tens to hundreds of times more productive rather than copying foreign methods.
  • Sankar asserts that a shift toward financial engineering over real engineering, including finance-oriented leadership choices, contributed to industrial decline in firms such as Intel and Boeing.
  • Sankar claims Europe has created zero companies from scratch in the last 50 years worth more than 100 billion euros, while the U.S. has created multiple trillion-dollar companies from scratch in that period.

Defense Reform Tempo And Internal Resistance Dynamics

  • Sankar asserts that AI and modern tooling enable domain experts in uniform to build operational software quickly, shifting innovation from PowerPoint proposals to empirical prototypes.
  • Shyam Sankar asserts that the U.S. Department of Defense has experienced more change in the last year than in the prior 19 years.
  • Sankar asserts that innovation in defense bureaucracies requires leadership that protects and empowers internal nonconformists ('heretics').
  • Sankar asserts that large institutions often move from zero-to-one innovation into scaling prematurely, and can learn more about correct scaling paths from startups than from other large enterprises.
  • Sankar asserts continuity of specialized leadership (as exemplified by Rickover's long tenure running naval reactors) can be critical for delivering exceptionally safe, complex systems.
  • Sankar claims that Col. Drew Cukor's AI efforts related to Project Maven faced intense internal pushback, including IG threats and a baseless criminal-investigation allegation.

Defense Industrial Base Structure And Monopsony-Driven Stagnation

  • Sankar argues that the primary harm of post-Cold War defense consolidation was conformity and financialization more than reduced competition among vendors.
  • Sankar claims the share of spending on major weapons systems going to dedicated defense specialists shifted from about 6% in 1989 to about 86% today.
  • Sankar asserts that DoD monopsony constraints isolate defense suppliers from commercial competition, producing specialized but non-competitive 'Galapagos' firms.
  • Sankar claims that when Palantir started there was effectively no 'front door' for outside companies into DoD beyond In-Q-Tel in the intelligence community.
  • Sankar asserts that defense innovation historically came more from competition and disagreement inside government services than from head-to-head competition among defense firms.

Ai Value Capture And Enterprise Software Repricing Toward Differentiation

  • Sankar asserts that AI will pressure 'beta' software that makes organizations more similar, while 'alpha' software that expresses competitive differentiation will gain value.
  • Sankar argues that COVID exposed many large enterprise software investments as failing to deliver resilience or operational value, while collaboration tools were the standout value driver.
  • Sankar argues that applying AI to business should emphasize human-AI teaming to make top performers more productive rather than focusing on replacing workers.
  • Sankar predicts that models in the AI stack will remain commoditized under pressure while durable value accrues primarily to chips and AI infrastructure.
  • Sankar reports that CEOs he works with are primarily asking how to dominate their industries rather than how to use AI to fire large numbers of employees.

Watchlist

  • Sankar predicts that over the next 2–10 years entertainment will shift toward more optimistic, pro-American stories with aspirational heroes rather than cynical antiheroes.

Unknowns

  • What objective indicators support the claim that DoD has changed more in the last year than the prior 19 (e.g., contracting cycle time, number of software programs reaching programs-of-record, policy changes implemented vs announced)?
  • Are the spending-share figures (6% in 1989 to 86% today going to dedicated defense specialists) accurate under a consistent definition of categories and 'major weapons systems'?
  • How common is retaliatory pushback like the claimed Maven case, and what institutional mechanisms (if any) measurably reduce it?
  • To what extent are domain experts in uniform actually building and deploying operational software with AI/tooling, and what fraction of those prototypes scale into sustained, supported capabilities?
  • Did Israel's post-October 7 reservist mobilization produce measurable modernization outputs (systems deployed, cycle times reduced, operational effects), and what were the enabling conditions?

Investor overlay

Read-throughs

  • Defense software and AI tooling could shift procurement toward faster prototyping and deployment, benefiting vendors and platforms that enable rapid iteration, integration, and secure delivery over long cycle major programs.
  • Institutional throughput and governance could become the binding constraint in defense modernization, increasing the importance of mechanisms that protect internal dissent, reduce retaliation against innovators, and maintain leadership continuity for complex software programs.
  • AI may commoditize generic enterprise workflow features while raising the value of differentiation enabling software that expresses unique operational advantage, shifting budget scrutiny toward measurable performance gains rather than feature parity.

What would confirm

  • Shorter contracting and delivery cycle times for defense software, increased number of prototypes reaching sustained programs of record, and implemented policy changes that reduce barriers to fielding and updating operational code.
  • Documented, repeatable cases of domain experts in uniform building deployable operational software with modern tooling, plus clear pathways for prototypes to scale into supported capabilities with maintenance, security, and ownership defined.
  • Evidence of enterprise software spend repricing toward differentiation, such as declining willingness to pay for generic workflow suites and increased spend on systems tied to measurable competitive or operational outcomes.

What would kill

  • No measurable improvement in defense contracting speed, deployment frequency, or software programs transitioning into sustained capabilities, despite continued announcements of reform or AI adoption initiatives.
  • Persistent or increasing retaliatory dynamics against internal innovators, with limited institutional mechanisms to protect dissent and a pattern of stalled pilots that never scale into maintained operational systems.
  • Customer behavior indicates generic enterprise workflow software retains pricing power and budget priority, while differentiation oriented tools fail to demonstrate ROI or are deprioritized as commoditized features meet needs.

Sources