Resolves as YES if there is strong evidence that before January 1st 2040 the global innovative drug market has functionally bifurcated into at least two major regional “stacks” (e.g., US/EU as one region, China as another; India may count as a third), such that each region routinely develops and approves distinct, non‑identical, domestically discovered (AI‑assisted) drugs for the same first‑line indications, without a unified global launch.

Operational criteria (any two of A–C must be met):

• A. Region‑specific analogs: In ≥3 major therapeutic areas (e.g., oncology, metabolic/obesity, autoimmune), ≥2 major regions each have an approved first‑line medicine for the same indication that:

  • is a different molecule/biologic (not a generic/biosimilar of the other);

  • was discovered/designated AI‑assisted by the sponsor (per regulator filings, company disclosures, or peer‑reviewed sources); and

  • remains unapproved in the other region(s) for ≥24 months after the first approval (no global “catch‑up” launch in that window).
    
    

• B. Localized R&D stacks: For each of those same ≥2 regions, at least two of the following hold at launch for the drugs counted in A:
  (i) domestic AI discovery platform used;
  (ii) ≥50% of pivotal trial sites in‑region and pivotal data primarily stored/processed in‑region;
  (iii) API and drug‑product manufacturing in‑region;
  (iv) distinct, region‑specific pricing & access decisions independent of ex‑region comparators.
  
  

• C. Policy‑driven separation: Credible evidence that trade, security, or data‑localization rules (e.g., tariffs, procurement restrictions, cross‑border data limits) were a material reason sponsors chose not to seek (or delayed seeking) cross‑region approval for at least two of the drugs counted in A.
  
  

What counts / does not:

• Counts: Different small molecules/biologics for the same disease (e.g., region‑A GLP‑1/GIP agent vs region‑B GLP‑1/GCG agent) discovered with AI tools and launched separately; distinct PD‑(x) antibodies with separate originators; region‑specific first‑line targeted therapies.
  
  

• Does not count: Generics, biosimilars, mere brand/label differences; secondary indications; co‑promotes/licensing of the same molecule; vaccines/diagnostics only; differences that are solely reimbursement timing without distinct molecules.
  
  

Adjudication notes:

• Acceptable evidence: regulator databases (FDA/EMA/NMPA, etc.), company filings, respected trade press, and peer‑reviewed literature.
  
  

• If evidence is mixed, default to NO unless the A–C thresholds are clearly met.
  
  

• This market does not require showing that one region is cheaper or higher quality overall; it’s about structural parallelism and analogous (but non‑identical) medicines replacing today’s default of a single global launch.