MY-DC Tactical Atlas — 2026-06-05

Top 3 Signals This Cycle

1. DDSP Closes RM1.33B Green Financing for Johor AI Datacenter

Why it matters: First major verified green-financed deal in the JS-SEZ corridor. 45MW liquid-cooled AI facility in Sedenak Tech Park. Capital markets have moved from announcements to financial closes — bankability is now proven for Malaysian AI DCs. Signals that liquid-cooling + green financing is the template lenders accept. Voices: China-B, China-C, Western-A, Western-B (6 mentions across cluster, all 4 voices aligned) Date: 2026-05-31 to 2026-06-02 Horizon: Quarter Risk Grade: A Confidence: 5/5

2. TNB RP4 Tariff Overhaul — 14% Base Rate Hike with Voltage-Tiered Pricing

Why it matters: Base tariff jumped ~14% to 45.62 sen/kWh with voltage-based structure specifically penalizing heavy DC loads. Peak-hour penalty tier under Energy Commission discussion. Power economics for Malaysian DCs just shifted structurally — every pro forma built before Q1 2026 is now stale. Combined with Batam undercut (Signal 3), Johor's cost moat is eroding. Voices: China-C, Western-A, Western-B, Western-C Date: 2026-06-04 Horizon: Quarter Risk Grade: B Confidence: 5/5

3. Batam Undercuts Johor by 22% on Power — 5-Year Tariff Freeze at $0.045/kWh

Why it matters: Indonesia just made the most aggressive policy move in SEA DC capture. Batam at $0.045/kWh vs Johor $0.058/kWh — Singapore-origin operators are already making relocation inquiries. This is the first credible regional threat to Malaysia's overflow thesis. If Batam grid + connectivity holds, Johor pipeline slows in 12-18 months. Voices: China-C, Western-A, Western-B, Western-C Date: 2026-06-01 Horizon: Quarter Risk Grade: B Confidence: 4/5

Top 3 Blindspots

1. ESA Overcommitment — 7.1 GW Signed vs 47% Historical Utilization (Western-only)

Gap: Only Western-A, B, C flagged this. China voices silent. TNB has 49 signed Electricity Supply Agreements totaling 7.1 GW committed DC demand, but historical utilization sits at 47% of declared maximum. Two readings: (a) massive latent demand wave coming → future grid crisis, or (b) significant overclaiming/demand slippage → ESA queue is softer than headline. Either way, the published pipeline is misleading. Voices: Western-A, Western-B, Western-C (China camp absent) Date: undated_estimate — DATE_UNKNOWN, verify before action Risk Grade: B Confidence: 3/5

2. Grid Lead-Time Lag — Cyberjaya 275kV Substation Not Live Until Q4 2026 (Western-B only)

Gap: Only one voice flagged the multi-year lag between TNB's RM43bn grid plan and actual capacity relief. Critical HV infrastructure has 24-36 month build cycles. Capital is flowing in faster than copper can be hung. Implication: any 2026-2027 power connection promise should be verified against specific substation commissioning dates, not framework commitments. Voices: Western-B (solo) Date: undated_estimate Risk Grade: A Confidence: 3/5

3. Johor Water Constraint — 142 MLD Supply vs 808 MLD DC Demand (Western-B only)

Gap: Only Western-B surfaced the water math. Johor's water infrastructure supplies 142 megaliters/day against datacenter demand of 808 MLD — a 5.7x deficit. Concurrently TNB's renewable mix for DCs fell to 18% in May 2026. This is the ESG/operational bomb under the Johor thesis that the China camp and most Western voices are not pricing. Voices: Western-B (solo) Date: 2026-05 (May 2026) Risk Grade: A Confidence: 3/5

Top 3 Contrarians

1. Equinix Cyberjaya 45MW IBX — Enterprise Demand Reaffirmed Outside Johor

Contrarian read: While narrative is Johor-Johor-Johor, Equinix (the interconnection bellwether) is planting a 45MW phased build in Cyberjaya starting Q3 2026. Suggests enterprise + interconnection demand is bifurcating from hyperscale AI training demand. For Vincent's Perak plot, Cyberjaya's revival matters more than Johor's saturation. Voices: Western-B, Western-C (2 voices) Date: 2026 Q3 build start Horizon: Year Risk Grade: C Confidence: 3/5

2. Federal-State Regulatory Bimodality — Not Just Slow Bureaucracy

Contrarian read: Federal Energy Minister explicitly rejected moratorium calls (2026-06-04) one day after Johor CM pushed for environmental pause (2026-06-02). This is not bureaucratic friction — it's a structural federal-state power struggle over resource allocation. Outcomes are bimodal: either federal wins and Johor pipeline accelerates, or state wins and a de facto moratorium emerges through permit slow-walking. Plan for both tails. Voices: Western-A, Western-B, Western-C, China-C Date: 2026-06-02 to 2026-06-04 Horizon: Month Risk Grade: A Confidence: 4/5

3. China-B Cross-Border IP Transfer Frame on DDSP Deal

Contrarian read: China-B uniquely framed the DDSP financing as "proving cross-border IP transfer" — meaning liquid-cooling tech and AI DC design IP is now flowing Singapore → Johor through capital structures. Western voices treated it as a financing event; China-B treats it as a tech transfer event. If correct, expect China-affiliated operators to replicate the template in 2026 H2. Voices: China-B (solo framing, confidence 0.95) Date: 2026-05-31 Horizon: Quarter Risk Grade: C Confidence: 3/5

Tracked Forecasts Update

FORECAST: Johor Chief Minister's environmental pause pressure translates into either a permit slowdown or a formal state-level review of new DC approvals, despite federal rejection of moratorium | HORIZON: month | VERIFY_AFTER: 2026-07-05

FORECAST: At least 2 more green-financed AI DC deals (>$200M each) close in JS-SEZ corridor following DDSP template | HORIZON: quarter | VERIFY_AFTER: 2026-09-05

FORECAST: TNB RP4 peak-hour penalty tier for heavy DC loads gets formalized by Energy Commission, adding 8-15% to effective DC power costs | HORIZON: quarter | VERIFY_AFTER: 2026-09-05

FORECAST: At least one Singapore-origin DC operator publicly announces Batam site selection over Johor citing tariff differential | HORIZON: quarter | VERIFY_AFTER: 2026-09-05

FORECAST: Equinix Cyberjaya 45MW IBX Phase 1 breaks ground on schedule (Q3 2026), validating Cyberjaya as enterprise alternative to Johor | HORIZON: quarter | VERIFY_AFTER: 2026-10-05

FORECAST: Johor water supply-demand gap triggers either a state-imposed water allocation cap on new DCs OR a forced shift to closed-loop/seawater cooling mandates | HORIZON: year | VERIFY_AFTER: 2027-06-05

FORECAST: TNB's 7.1 GW signed ESA portfolio shows <60% actual energization by mid-2027, confirming overclaiming thesis | HORIZON: year | VERIFY_AFTER: 2027-06-05

Opportunity Map — Vincent-Specific

Signal 1 Action — DDSP Green Financing Template

90-day action: Engage one Malaysian green-finance advisor (CIMB, Maybank Islamic, or RHB Sustainable Finance desk) to scope whether Alor Pongsu plot can be packaged as a green-financed AI DC site (liquid-cooled, RE-backed PPA). Get a term sheet indicative, not a commitment. Estimated RM cost: RM 40K-80K (advisory retainer + feasibility memo) 30-day disprove test: If no Malaysian bank's sustainable finance desk will even take an exploratory meeting on a Perak-located AI DC site within 30 days, the green-financing thesis does not extend outside JS-SEZ — kill it.

Signal 2 Action — TNB RP4 Tariff Exposure Mapping

90-day action: Commission a TNB tariff scenario model for Alor Pongsu under three RP4 outcomes (base 45.62 sen, +peak penalty, +voltage tier). Cross-reference Perak substation capacity at the nearest 132kV/275kV node — get the actual queue position and lead time in writing from TNB Grid Division. Estimated RM cost: RM 25K-50K (independent power consultant + TNB engagement fees) 30-day disprove test: If the nearest viable HV substation has >36 month connection lead time OR is already fully subscribed by existing ESAs, the Perak plot's DC thesis is dead — pivot to logistics/industrial use.

Signal 3 Action — Batam Competitive Benchmark

90-day action: Send one analyst (or hire a Singapore-based DC consultant for 2 weeks) to Batam to map actual operational readiness: grid reliability, subsea cable connectivity to SG, permitting reality vs the $0.045 headline tariff. The threat to Vincent's Malaysian thesis is not the tariff number — it's whether Batam can actually execute. Estimated RM cost: RM 30K-60K (consultant + travel) 30-day disprove test: If Batam grid SAIDI/SAIFI numbers, subsea cable capacity, or permitting timelines are materially worse than headlines suggest, the 22% tariff arbitrage will not trigger relocation — Johor (and by extension Perak as overflow) holds. If Batam checks out, Vincent must reposition Alor Pongsu away from hyperscale and toward domestic/enterprise demand (Equinix-Cyberjaya pattern).

Confidence & Coverage Note

Coverage: 10 topics completed across MY operators, hyperscalers, SG overflow, TNB grid, water/cooling, land/zoning, regulation, risk, Perak-specific, SEA comparative. 212 signals, 83 clusters, 4 consensus / 1 camp-split / 61 outlier.

Confidence caveats: Consensus cluster is heavily Western-weighted (3 of 4 consensus clusters have Western-majority composition; China-A and China-D are conspicuously silent on tariff/regulatory mechanics). China-camp is strong on capital flows (DDSP) and weak on grid/water/regulatory granularity — this creates a real blindspot on operational execution risk.

Date integrity: Core signals (DDSP, RP4, Batam, federal-state split) all dated within last 14 days — fresh. ESA overcommitment, water deficit, and grid substation lead-time signals are flagged DATE_UNKNOWN or May 2026 — verify before acting on these.

Perak-specific gap: Of 10 topics completed, Perak-specific signals appeared in only 1 — meaning Vincent's plot is being analyzed by analogy to Johor/Cyberjaya, not on its own data. Next cycle should prioritize TNB Perak substation queue, Perak state DC policy, and Perak water authority capacity directly.