DigestAlpha

[MS2 Capital @ms2capital]

🧠 AI MEMORY SHORTAGE (60 SEC BREAKDOWN)

We modeled supply vs demand for HBM4 = the memory every AI chip is bolted to. Only 3 companies on earth make it - $MU is one of them.

▪️Demand > supply every quarter through 2028
▪️Worst gap: late 2027 (only ~70% of orders filled)
▪️Bottleneck isn't fabs, it's stacking machines
▪️16 perfect layers per chip (1 crooked layer= trash)

📡SK hynix just panic-ordered more = the signal.

▪️Price path: $16.60 → $30+/GB. Bull: $53
▪️Buyers not bluffing: each $16.60/memory ~$29/yr AI revenue
▪️Micron at $900 looks expensive, but trades under 8x our 2028 estimate.
▪️Street EPS went $12 → $108 in a year.

The market believes the money, but it just doesn't believe it lasts - our model says it lasts through 2027, but 2028 is the fight.

🔔4 smoke alarms to get us out early:

▪️memory prices below model
▪️machine orders slowing
▪️AI revenue per GB rolling over
▪️+94% YoY industry growth peaking

Every boom funds the overbuild that ends it - our alarms are built to ring 2 quarters before the top🚀

[MS2 Capital @ms2capital]

$MU | HBM4 Demand

Memory is eating the rack - Bernstein's new Vera Rubin math👇

▪️ $3.2M memory + storage per $9.1M rack ~35% of BOM vs 26% on stale marks

▪️ HBM4 alone: $0.34M → $1.09M per rack at $53/GB

Micron isn't a memory cycle trade anymore, it's a bottleneck stock🚀

[Joey @ NS @joeysantoro]

This plus retardmaxxing implies being an insecure intellectual is the worst quadrant to land in

[Photon Capital @PhotonCap]

I have received a lot of DMs about SemiAnalysis’ CPO delay discussion and the optics sell-off that followed.

After looking into it, I am preparing a full article.

But the conclusion is simple: I bought optical stocks yesterday. I did not sell $SIVEF, $AAOI, $COHR, $LITE, $CIEN, or my broader optics exposure.

[MS2 Capital @ms2capital]

$AAOI | CPO Delay Headline

Yesterday @AppliedOpto dropped 17% on a CPO delay report. The market sold the wrong product line. Quick breakdown:

What actually happened:

Semi analysts flagged delays in co-packaged optics (CPO) adoption timelines -> entire optical complex sold off:

• $LITE -8%
• $COHR -11%,
• $CIEN -7%

Why the selloff didn't make sense for AAOI:

• CPO is NEXT-gen architecture. AAOI's business today is 800G pluggable transceivers for the current hyperscaler build.

• A CPO delay doesn't break the 800G story - it extends it.

• Rosenblatt's $140 target was built on pluggables, not CPO optionality.

What the fundamentals say:

• TTM revenue: $507M, +64% YoY
• FY2025 revenue +83%, accelerating off a +15% 2024
• InP laser fab capacity expanding 350% by 2027
• Hyperscaler 800G lead times extending, not contracting

The real risk (watch closely):

• Competition - LITE, COHR, and InnoLight are all better-capitalized

• A preferred-vendor loss caps the revenue ceiling. That's the bear case, not CPO

Why we stay bullish on photonics:

• AI datacenter buildout is bandwidth-constrained, and bandwidth = optics

• Record current demand + a forward-architecture scare = sector-wide dislocation

• Selling today's cash flows over tomorrow's roadmap is how entries get made🚀

[MS2 Capital @ms2capital]

🛰️ $AAOI | Applied Optoelectronics

The photonics market is still underpriced

Why photonics matters👇

AI clusters have outgrown copper. The fix is light- photonics converts electrons → photons to move data at GPU speed. It's a full stack (materials → packaging → optical engines → deployed networks), and bandwidth is now the bottleneck for every AI buildout.

📊See stack framework from @crux_capital_

Where AAOI sits: top of the stack - the 800G/1.6T transceivers that physically wire GPUs together.

• Q1'26 rev $151M, +51% YoY — 5th straight record quarter

• Datacenter rev $81M, +154% YoY - first volume 800G shipments to a hyperscaler

• FY'26 guide raised to >$1.1B, more than 2x last year

• $324M+ in 800G/1.6T orders booked; first 1.6T order in hand

• Capacity scaling ~100K → 500K units/mo by year-end

The $NVDA angle 🟢

• Every NVDA GPU needs optical interconnect - AAOI builds exactly that plumbing

• NVDA's ~$4–6.5B optical bet (Lumentum, Coherent, Corning) confirms bandwidth = the new gold

• AAOI is the one major transceiver name without an NVDA equity check yet → asymmetric upside if that changes

Photonics is the AI backbone. At ~$14.6B, AAOI isn't priced like it🚀

[MS2 Capital @ms2capital]

$ZEC | Steelman Argument

Most chains can not prove a bug wasn't exploited, but Zcash is one of the few built to - here's how the foundation could salvage this👇

🔒Turnstiles: shielded pools can't release more value than what entered, so even if a soundness bug minted fake coins, that value is trapped - it can't be cashed out to transparent ZEC without the network rejecting it.

📊Supply reconciliation: total ZEC is auditable against the emission curve. Supply matches issuance → no inflation was ever realized.

🔍Forensic search: scan all chain history for the malformed-proof signature an exploit requires - show it never appeared in a single block.

🛠Patch-first disclosure: fixed in a network upgrade before the reveal = a closed hole, not an open one.

✅Independent audit: outside cryptographers co-sign the analysis. "Trust us" becomes "verified."

📜Precedent: In 2018–19, ECC found a counterfeiting flaw, patched it in Sapling, then disclosed the finding - credibly proved no exploitation via the turnstiles. ZEC survived and recovered.

If they run that playbook:

• Violent relief rally as the "this is fatal" discount unwinds

• $ZCSH double tailwind - NAV recovers and its current ~24% discount to NAV compresses back toward ~11% or less

• Narrative flips to "the turnstile design did its job" - arguably stronger than before.

⚠️ The catch: this only works if it's a counterfeiting/soundness bug (turnstile-auditable). If it's a privacy/deanonymization flaw, you can't prove a clean negative - different, murkier story.

[MS2 Capital @ms2capital]

$BTC | Bear Market Bottom Signals

While @MicroStrategy and @BitMNR are deep underwater, these are unrealized marks on leveraged treasury vehicles - they don't represent the entire asset class:

🔴MSTR unrealized loss: –$10.9B
🔴BMNR unrealized loss: –$9.3B

On the surface it looks bleak, but @CoinDesk reminds us to check what onchain is signaling:

🟢10.5M BTC now sit at an unrealized loss vs. 9.8M in profit - supply-in-loss > above supply-in-profit

🟢This crossover has marked every major bear bottom: 2015, 2019, 2020, 2022

🟢 Price tagged the 200-week MA at ~$61.3K - the support level touched at every cycle low

The treasury equities are the volatility expression, but the onchain cost-basis flush is the signal of capitulation and what marks cycle bottoms🚀

[MS2 Capital @ms2capital]

$VVV x $VIRTUAL - Agentic Finance on @base🚀

@AskVenice now powers private inference for every @virtuals_io agent.

🔥 $VIRTUAL: $70.6M fees all-time (100% on Base)
🔥 $VVV: $3.83M/yr take, 32.5% of supply burned

Agents stake → mint $DIEM → VVV leaves supply. 26.4% already locked.

$400K in inference credits: lower bar for ideas → live agent.

Same chain + same demand - we'll be watching DIEM mints...👀

[MS2 Capital @ms2capital]

The Gamma Squeeze | $PURR

It's no secret @Globalflows has been leading the charge on PURR for his community. The only liquid proxy for HYPE ripped to $11.62 this AM.

Over the weekend, he shared an even bolder take: PURR could see a 2021 $GME-style gamma squeeze.

The setup:

• Largest volume day EVER
• Call OI vertical at ~98K contracts
• ~9.8M shares of delta on dealer books - short gamma means every tick up can force hedge buying

The potential fuse:

🔥 Strikes are capped too low - once new higher strikes list, gamma ramp resets and reloads.

🔥 If dealers are short these calls, NAV premium + $HYPE token demand + thin float all gap it higher.

⚠️Thin liquidity cuts both ways, and NAV premiums snap back hard.

The event is a potential asymmetric lottery ticket - size appropriately and watch the strike chain🚀

[Capital Flows @Globalflows]

Hyperliquid Strategies ($PURR ) will have a gamma squeeze in the next 60 trading days (similar to GameStop)🧵👇

$PURR just had its largest day of trading volume, indicating how aggressively investors are establishing positions into the regulatory change for Hyperliquid

On top of this, call open interest for $PURR is surging, as traders buy the OTM tails. Watch very closely because once more OTM calls get listed, it will almost certainly cause a gamma squeeze. Right now, $PURR is the only liquid location to buy OTM calls on Hyperliquid, squeezing into the regulatory acceptance.

There is a massive problem with the calls right now, though. The strikes aren't listed very high. I'll explain this in the next tweet below for you.

[MS2 Capital @ms2capital]

Majors' Market Signals | $BTC + $HYPE

@cburniske has always shared reliable insights on broader markets, but he may be overlooking what the majors are actually signaling:

1️⃣ CEO of ICE (owns NYSE) called @HyperliquidX "bigger than NASDAQ"

2️⃣ CFTC just greenlit 24/7 trading, perps, and BTC collateral

3️⃣ Same guidance powers the $MSTR flywheel + unlocks $STRC as BTC-backed digital credit

TradFi isn't waiting for majors to rotate; they're building new rails, and confirmation comes after the move, not before.🚀

[Chris Burniske @cburniske]

I don’t see a new paradigm where majors are rotated out of, and select up-and-comers can sustain their swim against an outgoing tide for long. Instead, I think majors are telling us something, and people are finding reasons not to listen.

[MS2 Capital @ms2capital]

ZEC Spot Discount | $ZCSH

@Grayscale ZCSH product is trading at a ~8% discount to NAV right now.

• ZEC: $571.21
• ZEC per share: 0.0807
• NAV: $46.08
• Price: $42.41
• Discount: −7.97%

Two paths to upside:

1️⃣ ZEC pumps
2️⃣ Discount closes

If @RaoulGMI is right about 5-10% of $BTC market cap and we're not headed to Goblin town, you can scoop $ZEC at a decent discount🫡

[MS2 Capital @ms2capital]

ETF Inflows | $HYPE - The Double Bid

For the first time, a major has two programmatic buyers on the same float:

• Assistance Fund: buys with protocol revenue
• ETFs: $THYP and $BHYP buy with TradFi flows

In just 10 days, here's their impact:

• Inflows > $101M and counting
• 1.91M HYPE locked in ETF baskets
• 0.57% of float absorbed
• At current pace → 20%+ behind glass in 12 months

Every other major has ETF demand or a buyback - @HyperliquidX has BOTH... and they're accelerating🚀

[MS2 Capital @ms2capital]

$NEAR | 1.2% -> 10% Ethereum Market Cap

NEAR mcap: $3.01B (1.2% of ETH)
ETH mcap: $255.6B

NEAR trading~1.2% of $ETH's market cap is structurally mispriced, here's why:

🔹100% of NEAR Intents fees → open-market token buybacks (live since Feb 23)

🔹$79M/day volume → 45% of the $177M deflation threshold, accelerating

🔹0 → 42% confidential transactions in 6 weeks according to @ilblackdragon

🔹 150+ assets, 35+ chains - the only cross-chain confidential settlement layer at scale

🔹Payments, payroll, treasury, agent-to-agent flows +49% this week + the privacy category premium isn't even priced in yet..

This isn't a "flip ETH" trade - it's an asymmetric bet that confidential cross-chain settlement becomes a distinct category - NEAR owns this primitive.

TEE-based privacy unlocks a TAM that Ethereum structurally cannot serve natively - @NEARProtocol capturing 10% of ETH's mcap = 9x re-rate to $20+🚀

[MS2 Capital @ms2capital]

Bears are Wrong | Easy Mode Playbook

@TaikiMaeda2 dropped his "bears are wrong" thesis and revealed his Holy Trinity for the cycle ahead:

🔹 $BTC Power - Saylor's reflexive STRC loop.

🔹 $ZEC Wisdom - privacy + quantum hedge.

🔹 $HYPE Courage - revenues to holders + no VCs.

Fear & Greed printed its lowest reading in crypto history at $66K - checkout his video below + summon the green candle therapy🚀

[MS2 Capital @ms2capital]

$ZEC l Beras Look Cooked

Funding is split on @GMX_IO🫐

🟢Longs receive +0.0020%/hr (positive funding)
🔴Shorts pay -0.0019%/hr plus a -0.0114%/hr borrow fee

Ooof might be time to close those shorts beras🚀

[Barry Silbert @BarrySilbert]

Fantastic discussion re Zcash. Great intro to $ZEC and investment case

[CoinGecko @coingecko]

NEWS: Robinhood lists $VVV, causing it to pump 10.2%.