Author · brief 2026-06-22

@lokoyacap lokoyacap

Supply-chain-deep AI compute analyst modeling hyperscaler capex and chip financing

Posts original, reasoning-heavy threads on the AI datacenter

trader score
-1.46
hit rate
50%
mean α
-0.61%
signals 14d
27

Grade = how their written analysis reads (A best). Trader score = how their last-20 timestamped calls performed vs SPY. · Analyst brief as of 2026-06-22.

Their picks, scored

Across their last 20 scored bets: 50% hit rate, -0.61% mean alpha, trader score -1.46. Their last-14d mentions, direction-adjusted, have moved -4.1% since posting (mean over 11 mentions with price data).

AI capacity, model security, and selective infrastructure confidence

lokoyacap focused on AI infrastructure constraints, model-security politics, and a small number of company-specific data points. The clearest ticker concentration is MSFT, where the author reported both Azure capacity prioritization and OpenAI concern around distillation controls. There are no directional calls or position disclosures, so the window reads more like news-driven thematic monitoring than explicit trading.

Themes3analyst read · 2026-06-22
AI model protection and distillation risk
mixedfading13 signals
⚠ 77% of theme signals are GOOGL — flag pump risk
Cloud capacity demand and constrained earnings
bullconsistent3 signals
⚠ 100% of theme signals are MSFT — flag pump risk
Semiconductor and AI infrastructure supply chain
neutralconsistent1 signals
⚠ 100% of theme signals are ASML — flag pump risk
Direction this week

No CALL_DIRECTIONAL or POSITION_DISCLOSURE signals appeared, so there is no evidence of adds, trims, exits, or a position flip. The week is concentrated in AI capacity and model-security news, especially MSFT, with a constructive capacity-scarcity framework offset by a negative distillation-control report. LBT appears as a single bullish confidence signal, but it is too isolated to anchor a separate theme under the minimum-signal rule.

Best hypotheses2their highest-scoring claims in our index
Alphabet’s massive AI infrastructure raise is justified because AI demand exceeds supply and Cloud/Gemini usage already
bullHIGH6 co-supporters
Large investors, managers, and market structure narratives reinforce Amazon as a core compounder worth owning through cy
bullMEDIUM7 co-supporters
Recent signals15receipts included
date (PT)tickerauthorsentwhat they saidsince thenreceipt
2026-07-02META@lokoyacap+0.35Suggests recent pundit takes on Meta may be completely wrong.
2026-07-02·@lokoyacap·Cites FT report that a social media company may boost AI capex to $145B this year.·
2026-07-01·@lokoyacap·Says GPU fleet is multi-generation asset; old GPUs retain value and new GPUs remain scarce.·
2026-07-01·@lokoyacap·Compute sales likely small slice at accretive rates, supporting more investment.·
2026-07-01·@lokoyacap·Agrees with scenario of high compute-market returns for sellers of compute.·
2026-07-01META@lokoyacap+0.75Says IaaS-like business would enable much higher Meta capex with less dilution.-4.9%
2026-07-01META@lokoyacap+0.65Says compute sale would fund Meta doing more and is a genius move if it happens.-4.9%
2026-06-30·@lokoyacap·Cites Ramp saying AI is not killing jobs, challenging the AI labor displacement narrative.·
2026-06-28·@lokoyacap·Praises analysis on healthy unit economics for suppliers, financiers and asset owners.·
2026-06-28·@lokoyacap·Skeptical comment about workforce cuts as token spend peaked, but no ticker named.·
2026-06-28·@lokoyacap·Says hyperscalers remain short supply at larger footprints, supporting AI infrastructure demand.·
2026-06-28·@lokoyacap·Says two major companies are still significantly short compute.·
2026-06-27·@lokoyacap·Says focus on SCA pricing ignores scale of remaining roughly 80%.·
2026-06-27·@lokoyacap·Cites source saying administration limits on Fable 5 could be lifted as soon as next week.·
2026-06-26GOOGL@lokoyacap+0.25Says AWS and GOOGL are among most competent data center operators at scale.+6.7%

Grade is our human read-worthiness rating; trader score is a rolling 20-bet hit-rate/alpha composite — different things, often disagreeing. “Since then” is direction-unaware in the table; the summary line above adjusts for which way they leaned.