Dominant share at leading nodes plus scaled packaging underpin a credible path to ~$250B sales by 2030. Policy risk and overseas dilution temper, but don’t derail, upside.

Owns the compute-capex flywheel. With 2nm/A16 leadership and scaled CoWoS/SoIC packaging, TSMC can compound AI/HPC manufacturing share into >$250B 2030 revenue while keeping trust, yield and distribution moats.

Compute supremacy requires wafer + packaging + yields at global scale; TSMC is the pivotal manufacturing platform for AI era despite geopolitics/capex drag.

Deterministic: Revenue path. 1) Rev_2030 assumed 250,000 (USD-m) on leading-edge share, AI packaging mix, N2/A16 ramps. 2) EV/Rev_2030 multiple = 10x (peer premium sustained by compute capacity + trust). EV_2030 = 2,500,000. 3) Current EV ≈ current market cap 1,220,000 + net debt (debt–cash) ≈ −48,800 → 1,171,200. 4) 2,500,000 / 1,171,200 ≈ 2.1x. Comps cross-check: TSM trades high-teens EV/Rev TTM; Intel/Samsung foundry lower; sustaining 10x into 2030 is optimistic but feasible with packaging-led margins and capacity scarcity.

Key risks: policy/tariff shifts (incl. tool/export curbs), overseas fab cost dilution, yield/throughput at N2 and advanced packaging scale, and customer concentration. Execution track record and balance sheet offset much of this.

Thesis intact; raise 2030 revenue potential with packaging mix and sustained AI/HPC orders, balanced by overseas margin dilution and tariff noise.

Maintain fast-growth potential; implied upside moderated slightly as current valuation rose and overseas dilution became clearer.