Stock Analysis · Bulls vs Bears

Gorilla Technology Group GRRR

🐂 Bulls vs Bears 🐻Report #212 Oct 2026
★ Willem owns this stock · 1.9% of the portfolio
72
high risk▼ −4 vs last
🐂 8vs12 🐻
Bears lead
15%
▲ Chance of the biggest tailwind the forecasts miss
45%
▼ Chance of the biggest headwind the forecasts miss
The verdict

Since the last report the evidence has moved slightly toward the bulls on delivery: Yotta hardware is now physically arriving. It has moved clearly toward the bears on the capital structure. The convertible indentures show a first downward reset on 2027-03-15, with floors of $6 and $8. Gross margin has fallen for four quarters in a row, and the project debt promised in May is still not closed. At $10.10 the stock…

🎙️ Bulls vs Bears podcast — Gorilla Technology Group
Alex and Sam debate every bull and bear argument · 13:54
0:0013:54
⤓ Download

Written by Claude (Anthropic’s AI) from public sources · Willem owns GRRR · AI can make mistakes — numbers in brackets point to the sources · Not financial advice

🐂 Bull arguments 8

  1. 1Hardware is finally landing: Yotta deliveries began late SeptemberMedium
  2. 2Market value is about half of consensus 2027 revenueMedium
  3. 3Loose debt terms: no covenants, nothing due before 2031Strong
  4. 4Nvidia anchors half the Yotta 2 offtake for four yearsMedium
  5. 5GPU rental prices rose in 2026; no glut yetMedium
  6. 6Revenue guidance raised three times in 2026; Q2 beatWeak
  7. 7Legacy cash: guarantees released, collections due, new Taiwan contractWeak
  8. 8A Fitch rating and wider coverage open the project-debt doorWeak

🐻 Bear arguments 12

  1. 1The March 2027 reset could nearly quadruple convertible sharesHigh
  2. 2Gross margin fell four quarters straight, to −4% in Q2High
  3. 3Promised project debt still unclosed; $228M already committedHigh
  4. 4Every major delivery date has slipped at least onceMedium
  5. 5Counterparties: levered Yotta, anonymous Batam buyer, landlord for saleHigh
  6. 6Interim numbers unaudited and unreviewed; backlog claims unreconciledMedium
  7. 7FY2025 profit and cash promises were missedMedium
  8. 8Capital is spreading into side bets: a US data centre, funds, stakesMedium
  9. 9Bigger, cheaper rivals are crowding Batam and IndiaMedium
  10. 10Supply-chain and export-control exposure via Supermicro and BISLow
  11. 11Heavy shorting and new resale supply on a falling chartMedium
  12. 12The Egyptian pound and slow receivables still weigh on the legacy bookLow

Bull arguments show how well they survive the devil’s advocate (strong / medium / weak); bear arguments show the damage they could do after the counter-argument (high / medium / low).

Bull case

🐂 Bull #1Hardware is finally landing: Yotta deliveries began late September

  • The point: The biggest doubt in report #1 was whether the GPU business exists at all. Physical delivery has now started, which is the first step towards rental billing.
  • Evidence:
    • Gorilla said the Yotta programme had moved "into physical execution" with equipment deliveries under way (2026-09-08) [19].
    • By 2026-10-05, the first of four B200 phases (640 servers, about 5,110 GPUs in total) was complete. The remaining three phases were due in early to mid-October. B300 deliveries were due to start in October in tranches of about 5,000 GPUs, ending by year-end [20].
    • Q2 revenue came partly from "earlier-than-expected delivery on contracted customer programmes" [2].
    • Gorilla has committed $227,965,267 to buy property and equipment, net of advances (2026-06-30) [1]. Together with 5-year depreciation and an assumed 20–25% residual value [13], this indicates that Gorilla will own the GPUs on its balance sheet.
  • Why it matters: Owned GPUs on five-year fixed-fee contracts are the whole thesis. Delivery moves it from a press release to an asset.
  • 😈 Devil's advocate: delivered is not billed.
    • Management itself says revenue depends on equipment being "delivered, configured, accepted, and placed into billing" [20]. No acceptance has been announced for any tranche.
    • The B300 programme was meant to be complete by 2026-09-30 [17]. It now ends around year-end at best [20].
    • Who legally owns the GPUs is still not stated in any document read. The $52.5M of H1 "hardware sales" were recognised as one-off sales, not rentals [1].
  • Verdict: Medium The physical progress is real and new since report #1, but acceptance and billing, which are what count, are still unproven.

🐂 Bull #2Market value is about half of consensus 2027 revenue

  • The point: if even part of the contracted revenue arrives, the stock is cheap on any sales multiple used for GPU clouds.
  • Evidence:
    • Price $10.10 on 2026-10-12, with 27.51M shares, gives a market value of about $278M [25,9].
    • Consensus (6 analysts, S&P) is FY2026 revenue $206.5M and FY2027 $574.9M [25].
    • Management guides FY2027 at $450–500M [2].
    • Analysts' average target is $32.83 (range $24–44) [25]. Zacks SCR uses 1.7x EV/sales for its $24 target [15].
    • Peers trade at roughly 5x (CoreWeave) and 7.7x (Nebius) forward EV/sales. These figures are from an undated summary [25].
  • Why it matters: the gap between price and the revenue path is large enough to pay for a lot of slippage.
  • 😈 Devil's advocate:
    • Price/sales on revenue that has not started is not a valuation; it is a probability.
    • Consensus ($574.9M) already sits about $100M above the top of management's own FY2027 range. It even builds in capacity that management has excluded [25,13].
    • Per share, the multiple depends on the March 2027 reset (Bear #1). Fully diluted at the $6/$8 floors, the share base would be about 61M, not 27.5M (estimate).
    • Peers own data centres and named hyperscaler customers. Gorilla does not.
  • Verdict: Medium It is cheap on the company's numbers, but those numbers and the share count are the two things in doubt.

🐂 Bull #3Loose debt terms: no covenants, nothing due before 2031

  • The point: the $232M of convertibles give Gorilla time. There is no maturity wall and no financial covenant to trip.
  • Evidence:
    • Both note series pay 7.50% and mature 2031-06-15. There are no financial covenants [6,7,9].
    • Gorilla can pay interest in shares, at 90% of the higher of the 5-day VWAP and the floor, provided a resale registration is effective [7].
    • Company redemption is allowed only from 2029-06-15. Holders can force repayment only on a change of control or a delisting [7].
    • Bank debt is small: short-term $9.3M plus Taiwan term loans at 2.5–3.0%, secured on land and buildings [1].
    • There is no going-concern language. The capital-management note uses standard wording only [1].
    • Cash was $179.4M at 2026-06-30 [1]. The July notes added about $120M net (reported) [9]. So cash is about $300M before H2 spending (estimate; no company update).
  • Why it matters: a small company buying hundreds of millions of dollars of GPUs is most likely to fail through a liquidity squeeze. These terms push that risk out several years.
  • 😈 Devil's advocate:
    • The flexibility is paid for in equity, through the resets and interest-in-shares (Bear #1).
    • A cross-default sits at only $5M of debt at the company or any significant subsidiary [7]. Once project loans exist at subsidiary level, a stumble there could reach the notes.
    • The notes forbid new debt senior to them without holder consent, and subsidiary debt is carved out [7]. So project debt has to sit in SPVs, which is where it has not yet closed (Bear #3).
  • Verdict: Strong There is no near-term solvency trigger. The cost shows up as dilution, not default.

🐂 Bull #4Nvidia anchors half the Yotta 2 offtake for four years

  • The point: the largest programme, Yotta 2 with 20,736 B300 GPUs, has the strongest possible end-user behind half of its capacity.
  • Evidence:
    • Yotta and Gorilla said Nvidia is expected to take about half of the offtake under a four-year commitment linked to DGX Cloud (Nvidia's own cloud service) (2026-04-30) [17].
    • The CEO said Nvidia "has taken one half of the offtake agreement with Yotta" (2026-05-07) [48].
    • Communications Today repeats the four-year DGX Cloud use (2026-10-05) [20].
    • Nvidia's own 10-Q shows AI-cloud and sovereign customers growing 138% year on year [45].
  • Why it matters: it cuts the risk that Yotta 2 sits idle, and Nvidia is the most creditworthy buyer in the chain.
  • 😈 Devil's advocate:
    • Nvidia's commitment is to Yotta, not to Gorilla. Gorilla's payer is still Yotta [17].
    • The commitment is four years; Gorilla's rental contract is five. Year five is uncovered.
    • The terms appear only in company press releases. Nvidia's filings do not name Yotta.
    • A four-year buyer of the newest chips may renegotiate when Vera Rubin, in production shipment since Nvidia's Q3 FY2027 [45], undercuts B300 pricing.
  • Verdict: Medium It is a real credit support, but one step removed and only partly overlapping in time.

🐂 Bull #5GPU rental prices rose in 2026; no glut yet

  • The point: the feared collapse in GPU rental prices has not happened. That makes Gorilla's fixed fees look safer and makes it easier to re-let capacity.
  • Evidence:
    • The Silicon Data B200 rental index was $5.62 per GPU-hour on 2026-09-01, up 27.6% year to date. H100 was $2.63 and H200 $3.29 (2026-09-07) [39].
    • One-year H100 contract rates rose from $1.70 (Oct 2025) to $2.35 (Mar 2026). This is SemiAnalysis data cited by a vendor [39].
    • E2E Networks (India) grew revenue 334% year on year, with a 75.2% EBITDA margin, and said growth came "primarily" from utilisation and capacity, not price [38].
  • Why it matters: if the market is tight, Gorilla's customers have little reason to walk away from contracts, and renewal pricing is supported.
  • 😈 Devil's advocate:
    • The B200 curve is backwardated: the 12-month forward rate is $5.03 against $5.62 spot, so the market expects about 10% lower prices [39].
    • India's IndiaAI programme rents subsidised GPUs at ₹65 per hour, "nearly one third of global average cost" [40]. That caps sovereign pricing.
    • The 2024–25 H100 collapse, from about $8 to under $3, shows how fast this can turn [39].
  • Verdict: Medium It supports the next 12–18 months, not the five-year contract life.

🐂 Bull #6Revenue guidance raised three times in 2026; Q2 beat

  • The point: management keeps raising the revenue bar and then clearing it.
  • Evidence:
    • FY2026 guidance went from $137–200M (2025-11-17) [12], to $160–200M (2026-05-27) [3], to at least $200M (2026-08-24) [2].
    • Q2 revenue of $50.1M beat the "at least $44M" pre-announcement [2].
    • Q3 guidance was raised to $48–50M from $36–40M [2].
    • FY2025 revenue of $101.4M landed in the $100–110M range [4].
  • Why it matters: revenue credibility supports the FY2027 guide of $450–500M.
  • 😈 Devil's advocate:
    • The beats came from one-off hardware sales at near-zero margin. Q2 had a gross loss (Bear #2) [1].
    • On the measures that matter (EBITDA, cash flow, gross margin), the FY2025 promises were missed (Bear #7).
    • The stock fell about 10% after the Q2 print, because the Q3 guide ($48–50M) and the FY2027 guide sat below analysts' $84.6M and $560M [29].
  • Verdict: Weak Hitting revenue with pass-through hardware says little about delivering rental margins. Report #1 rated this the same.

🐂 Bull #7Legacy cash: guarantees released, collections due, new Taiwan contract

  • The point: the old business still generates cash and contracts that reduce the need for new equity.
  • Evidence:
    • Restricted deposits fell from $5.30M to $45,933 as the Egypt guarantees were released [1].
    • Management expected more than $20M of collections in Sept–Oct and roughly breakeven operating cash flow for FY2026 (call summary) [13].
    • Q1 operating cash flow was +$6.6M [3].
    • The Taiwan Criminal Investigation Bureau AI contract (2026-09-22) is "tens of millions" of dollars, uses Nvidia B300 chips, and runs from Q4 2026 to Q4 2027 [24].
    • The Egypt revenue window now runs "through 2029", up from "through 2027" in the FY2025 note [1].
  • Why it matters: each dollar from legacy work is a dollar that does not need to come from dilution.
  • 😈 Devil's advocate:
    • H1 operating cash flow was −$4.3M, so Q2 alone was about −$10.9M [1,3].
    • Remaining unfulfilled long-term contracts fell from $101.3M to $72.7M in six months [1].
    • The credit-loss allowance rose more than fivefold, from $0.94M to $5.16M [1].
    • The extension to 2029 is unexplained.
  • Verdict: Weak It helps at the margin, but it cannot fund GPU capex. Its cash conversion is getting worse, not better.

🐂 Bull #8A Fitch rating and wider coverage open the project-debt door

  • The point: a first credit rating, plus a broader group of analysts, are the preconditions for the cheaper project debt that the model needs.
  • Evidence:
    • Fitch assigned a first-time B- rating with a Positive outlook on 2026-08-27. It expects project-SPV dividends to the parent "from 2029 or potentially earlier" [16].
    • New coverage: Compass Point initiated with Buy and a $44 target (2026-08-18) [28]. B. Riley initiated with Buy and $25 (2026-08-31). StoneX initiated with Buy and $25 (2026-09-29). Northland reiterated $40 (2026-10-02) [25,27].
  • Why it matters: rated, widely covered issuers get project finance on better terms.
  • 😈 Devil's advocate:
    • B- is deep junk. Fitch's projections assume delivery dates that have since slipped [16].
    • StoneX, and Benchmark (whose reported "Strong Buy" upgrade on 2026-09-29 appears in one aggregator only and is unverified), were the placement agents on both convertibles [7,26]. Their coverage is not independent.
    • Zacks SCR's research is issuer-paid [15].
    • Six weeks after the rating, no project debt has closed (Bear #3).
  • Verdict: Weak It is a necessary condition, not proof. The independence of part of the coverage is doubtful.

Dropped from report #1: "Optionality in Korat" is no longer a bull argument. Five months after the land purchase there is still no customer, financing or construction start. The October US data-centre LOI adds a second unfunded option (Bear #8). It now appears only as a tailwind consideration.

Bear case

🐻 Bear #1The March 2027 reset could nearly quadruple convertible shares

  • The point: the convertible notes reset their conversion price downward to the market price on fixed dates, starting 2027-03-15. At today's price that means many more shares, before the GPU business has even proved itself.
  • Evidence:
    • Each indenture, section 3.04(g), sets reset dates of 2027-03-15, 2027-12-31, 2028-09-30, 2029-06-30, 2030-03-15 and 2030-12-31. On each date, if the 5-day average price is below the conversion price, the new price is the greater of that average and the floor [6,7].
    • The floors are $6.00 for the June notes ($107M) and $8.00 for the July notes ($125M). This verifies the June floor that report #1 could not confirm [6,7].
    • Downward resets stop only once the later of 2027-03-15 and at least 50% of notes being converted has passed, or after a qualifying equity raise of at least 75% of principal [7].
    • Potential note shares (my arithmetic, principal ÷ price):
      • 9.10M at $25.48.
      • About 23.0M at a $10.10 reset (10.6M + 12.4M).
      • 33.46M at the floors (17.83M + 15.63M). That is 122% of the 27.51M shares outstanding [6,7,9].
    • Interest can also be paid in shares, at 90% of the price [7].
    • Authorised capital is 73.5M shares [9]. Outstanding shares, floor-case note shares, 9.58M warrants and 4.51M CVRs add up to about 75M before make-whole or interest shares (estimate from [5,9]).
  • Why it matters: below about $25 on 2027-03-15, every dollar of share-price weakness turns into more shares for noteholders. The reset happens whether or not Gorilla has delivered.
  • 🛡️ Counter-argument:
    • Resets only change the price at which holders may convert. With a reset price near the market price, holders still choose whether and when to convert.
    • If the stock recovers above $25.48 before 2027-03-15, the reset is moot. An upward reset caps at $31.85 [7].
    • A rally driven by Yotta billing in Q4 would reduce the damage directly.
  • Severity: High At $10.10, this is the single largest per-share risk, and its date is fixed.

🐻 Bear #2Gross margin fell four quarters straight, to −4% in Q2

  • The point: the business is getting less profitable at the gross level as it grows, the opposite of what the rental thesis needs.
  • Evidence:
    • Quarterly gross margin was about 37% in Q3 2025 ($9.9M on $26.5M), about 30% in Q4 2025, about 21% in Q1 2026 ($6.0M on $28.2M), and about −4% in Q2 2026 (about −$2.1M on $50.1M) [1,3,4,12].
    • H1 2026 revenue was $52.5M hardware (0% a year earlier) and $25.9M service. Security Convergence service revenue fell from $37.6M to $24.4M [1].
    • Memory costs rose 40–60% in four to five months (call summary) [13]. Contract DRAM prices rose about 93–98% quarter on quarter in Q1 2026 (TrendForce, via a summary) [41].
    • The 75% GPUaaS gross margin and the "40% plus" 2027 margin are not formal guidance [13].
  • Why it matters: if rental margins come in well below 75%, the FY2027 revenue still arrives but the cash to service $232M of notes and fund equity cheques does not.
  • 🛡️ Counter-argument: the decline is a mix effect. Mobilisation hardware is sold at cost ahead of rentals, and no rental revenue has started yet. Fixed-fee five-year contracts, once live, lock in the spread between fee and depreciation [13].
  • Severity: High The margin thesis is still untested, and the trend is still down.

🐻 Bear #3Promised project debt still unclosed; $228M already committed

  • The point: Gorilla has signed purchase commitments larger than its free cash, and the project loans meant to carry about 70% of the cost have not closed.
  • Evidence:
    • The commitment to buy property and equipment was $227,965,267 net of advances (2026-06-30) [1]. Estimated cash is about $300M after the July notes [1,9].
    • On the Q1 call (2026-05-27) the CFO said "the next announcement will be project-level financing closing", citing term sheets of $0.5–1B (summary) [14].
    • Batam "debt financing offers" cover about 70% of cost (2026-06-23) [21]. At Q2 there was "an offer" for 70%, and management said some initial offers were unfavourable (summary) [13].
    • No closing had been announced by 2026-10-12. EDGAR 6-Ks through 2026-10-08 cover only the notes [10].
  • Why it matters: without project debt, the equity portion becomes the whole cheque. That points to new equity or more convertibles, at a $10 share price.
  • 🛡️ Counter-argument:
    • The first 300 Batam servers are funded from prepayments and company cash [13]. Yotta 2's equity portion was what the June notes were for [5].
    • Lenders usually close once assets are accepted and billing, so the order of events is normal.
  • Severity: High It is five months since the "next announcement" promise, and the bill is already signed.

🐻 Bear #4Every major delivery date has slipped at least once

  • The point: the schedule has slipped repeatedly, and each slip moves revenue right while interest and depreciation keep running.
  • Evidence:
    • Yotta 1: first delivery "end of July" (Q1 call, 2026-05-27) [14] → shipped and deployed "by end August" (2026-07-21) [18] → first of four B200 phases complete in late September, the rest due early to mid-October (2026-10-05) [20].
    • Yotta 2: completion by 2026-09-30 (2026-04-30) [17] → Phase 1 complete by end-November (Q2 call) [13] → B300 deliveries October to year-end [20].
    • Batam: tranches in September and December 2026 (2026-06-23) [21] → first 300 servers by end-October with 20–30 days of testing [13]. No status found by 2026-10-12.
  • Why it matters: FY2027 guidance assumes full-year billing on these tranches. A one-quarter slip on about $425M of annualised GPUaaS revenue is about $106M of lost FY2027 revenue (estimate).
  • 🛡️ Counter-argument: each slip so far has been one to three months, not years. The hardware is now moving [19,20], and supply constraints on memory and GPUs hit the whole industry [41].
  • Severity: Medium High. The pattern is consistent, but each individual slip has been modest.

🐻 Bear #5Counterparties: levered Yotta, anonymous Batam buyer, landlord for sale

  • The point: Gorilla's five-year fees are only as good as the payers, and none of them is a transparent, strong credit.
  • Evidence:
    • Yotta:
      • Its MD said in July that the draft IPO papers would be filed with SEBI "within 2 weeks" (2026-07-02) [34]. The target then became October, with listing in Jan–Mar 2027 [35]. No DRHP had been found by 2026-10-12.
      • It sought up to $1.2B before the IPO and had "not secured all of the financing" [15].
      • Yotta 1 plus the first Yotta 2 batch are about $350M of Gorilla's FY2027 guidance (call summary) [13].
    • Batam:
      • The customer is still only a "high investment-grade global technology customer" [21].
      • Telkom's sale of about 70% of NeutraDC is at the binding-offer stage, targeted for end-2026, with no buyer named [36].
    • Concentration: three customers each exceeded 10% of H1 2026 revenue, and two accounted for more than 10% each of receivables plus contract assets [1].
  • Why it matters: if Yotta's IPO slips or fails, Gorilla's largest revenue line depends on a private, debt-funded operator.
  • 🛡️ Counter-argument:
    • Nvidia's offtake supports Yotta's revenue (Bull #4).
    • Yotta raised about $150M pre-IPO at about $3.9B [35].
    • NeutraDC's BTM-1 was "fully secured before launch", which makes it a sought-after asset rather than a distressed one [22].
  • Severity: High The Yotta IPO timetable has now slipped twice, and the Batam buyer is still unnamed.

🐻 Bear #6Interim numbers unaudited and unreviewed; backlog claims unreconciled

  • The point: the numbers investors are asked to trust are thinner than the claims built on them.
  • Evidence:
    • Marcum Asia was dismissed on 2026-04-30, two weeks after signing the FY2025 20-F. No reason was given beyond "a careful evaluation process" [11]. UHY was appointed.
    • The Q1 and H1 2026 statements are labelled "unaudited and unreviewed" [2,3].
    • The company claims "signed backlog over $5B" (Q1 call summary) [14]. IFRS unfulfilled long-term contracts are $72.7M [1].
    • Contract assets rose from $57.9M to $104.8M in six months [1].
    • Receivables 91–365 days overdue total $18.7M ($10.6M at 91–180 days and $8.1M at 181–365) [1].
    • The H1 figures conflict across filings. The F-3 shows cash of $179.1M and equity of $162.6M; the 6-K shows $179.4M and $170.9M [9,1].
    • The Q2 operating loss is about $6.1M from the IFRS figures, but "$2.2M" in the call summary [1,13].
  • Why it matters: the rental thesis rests on unpublished contracts. With no auditor review of interim numbers, a write-down of contract assets or receivables would come as a surprise.
  • 🛡️ Counter-argument:
    • Marcum reported no disagreements or reportable events [11].
    • Under IFRS 15, unrecognised GPUaaS contracts would not enter the backlog note until delivery, so the gap may be presentation, not substance.
    • EDGAR shows no SEC comment letters since 2024-02-05 [10].
  • Severity: Medium High. It is unchanged from report #1, except that the ballooning contract assets and the jump in credit-loss allowance make it sharper.

🐻 Bear #7FY2025 profit and cash promises were missed

  • The point: the revenue record looks fine; the record on everything else does not.
  • Evidence:
    • FY2025 guidance (reiterated 2025-08-14 and 2025-11-17) was adjusted EBITDA of $20–25M, adjusted net profit of $15–20M, positive operating cash flow, and a gross margin in the "40% range" [12].
    • Delivered: adjusted EBITDA $19.1M (below the range), operating cash flow −$28.9M, gross margin 33.4% [4].
    • In March the CFO said Gorilla "aims to be cash flow positive this year" [4]. H1 2026 operating cash flow was −$4.3M [1].
    • On the Q1 call, management guided to a FY2026 EBITDA margin "above last year, possibly 25–30%+" (summary) [14]. H1 adjusted EBITDA was −$14.6M [2].
  • Why it matters: the "75% rental margin" claim is a profit promise, and the profit promises have a poor record.
  • 🛡️ Counter-argument: FY2025 was disrupted by Egyptian pound devaluation (FX losses $21.0M) and FY2026 by the pivot itself. The misses are explainable rather than evidence of bad faith [4,5].
  • Severity: Medium

🐻 Bear #8Capital is spreading into side bets: a US data centre, funds, stakes

  • The point: a company that has not closed financing for its two signed programmes keeps opening new fronts.
  • Evidence:
    • US data centre: a non-binding LOI (2026-10-09) to buy a US data centre with about 6 MW today and a path to about 36 MW (roughly 18 months for the extra power). The plan is more than 100 MW of US capacity within 12–18 months. No price, financing or location was disclosed [23]. The stock closed −6.8% [25].
    • Korat: a 200 MW campus on bought land, with no customer [5].
    • Gorilla Tech Capital (formerly Shackleton Finance): bought for £1.0M. The target is $2–3B of funds under management by end-2027 [1,48].
    • Stakes: a 20.78% equity stake in Astrikos India, plus convertible notes, and a $5M SAFE with One Amazon [1].
  • Why it matters: each new front needs management time and equity, and the market is reading these moves as dilution signals.
  • 🛡️ Counter-argument:
    • The fund manager and the Astrikos stakes are small, under $5M in total [1].
    • The US LOI is non-binding, and a US site would reduce reliance on India and Indonesia.
  • Severity: Medium The amounts are small today, but the pattern matters given Bear #3.

🐻 Bear #9Bigger, cheaper rivals are crowding Batam and India

  • The point: Gorilla's niche of renting GPUs in places short of compute is attracting much larger, better-funded players.
  • Evidence:
    • Firmus (with Nvidia and DayOne) plans up to 170,000 accelerators on a 360 MW campus at Kabil, Batam, with operations from Q1 2027. It raised about $1.35B at a $5.5B valuation [37].
    • E2E Networks (India) runs about 5,100 GPUs at a 75.2% EBITDA margin and won a ₹10bn Blackwell order [38].
    • Yotta runs its own Shakti Cloud, so Gorilla's customer is also a competitor for end-users [51].
    • IndiaAI rents subsidised GPUs at ₹65 per hour [40].
    • Gorilla's scale: it owns no data centre, and property and equipment were only $29.4M at 2026-06-30 [1].
  • Why it matters: at contract renewal (2031), or if a customer defaults, Gorilla has no cost or scale advantage to re-let capacity.
  • 🛡️ Counter-argument: the contracts are five-year, fixed-fee and reportedly take-or-pay [16]. Competition matters at renewal, not before.
  • Severity: Medium It is mostly a 2029–31 risk, unless a counterparty walks away earlier.

🐻 Bear #10Supply-chain and export-control exposure via Supermicro and BIS

  • The point: Gorilla's hardware comes through Supermicro (about $2B of servers for Yotta [50]) into countries that US export policy watches.
  • Evidence:
    • Supermicro: on 2026-03-19, prosecutors in the Southern District of New York unsealed an indictment of Supermicro co-founder Wally Liaw and two others for allegedly diverting export-controlled servers to China. The company is not a defendant [42].
    • Supermicro's FY2026 10-K reports operating cash flow of −$6.81B and supply constraints on memory and GPUs. It says licences may be required for Southeast Asia [41].
    • BIS: guidance (about 2026-05-30) enforces licences for advanced chips going to China-headquartered entities anywhere in the world [43].
    • Draft rules: a 2025 draft rule for Malaysia and Thailand licences was reported but not finalised [43].
  • Why it matters: a licence hold or a shift of supplier could delay tranches that are already late.
  • 🛡️ Counter-argument: no 2026 rule imposes licences on India or Indonesia. Gorilla says it does not sell to or host in China [5]. Supermicro's independent investigation found no knowledge at current senior management [41].
  • Severity: Low Medium.

🐻 Bear #11Heavy shorting and new resale supply on a falling chart

  • The point: the stock's technical position amplifies bad news.
  • Evidence:
    • FINRA short interest was 4.84M shares (settlement 2026-09-30), about 17.6% of shares outstanding, with 10.6 days to cover. It peaked at 5.12M on 2026-08-14 [30].
    • Price $10.10, down about 29% in a month, against a 52-week range of $9.04–23.49 [25,31].
    • A prospectus supplement for resale by note holders is due 2026-10-15 [10].
    • The worst week in 18 months (−34%) followed the July notes [33].
    • Insiders made no open-market purchases in 2026. The only open-market trade found is a sale of 17,790 shares by director Medeiros in April [32].
  • Why it matters: convertible holders often hedge by shorting the stock. Each financing adds pressure, and lower prices feed directly into the March reset (Bear #1).
  • 🛡️ Counter-argument: high short interest with a hard catalyst (Yotta acceptance, Q3 results about 2026-11-18) can also produce sharp squeezes. The CEO holds 1.21M shares [32].
  • Severity: Medium

🐻 Bear #12The Egyptian pound and slow receivables still weigh on the legacy book

  • The point: the cash-generating legacy contract is priced in a weakening currency and paid slowly.
  • Evidence:
    • Fixed price of about EGP 8.4bn. EGP/USD averaged 44.18 in FY2024 and 49.28 in FY2025 [5].
    • FX losses were $21.0M (2025) and $18.9M (Q1 2026) [4,3].
    • The IMF's 8th and final review is due around mid-December 2026 [44].
    • The Egyptian government may terminate for delays of more than two months [5].
  • Why it matters: this is still the main source of gross profit until rentals start.
  • 🛡️ Counter-argument: the remaining value is only $72.7M, so the exposure is shrinking. The IMF completed the 7th review in July [44,1].
  • Severity: Low

Tailwinds NOT in the forecasts

Reference points:

  • Consensus FY2027 revenue $574.9M and EPS −$0.41 (S&P, 6 analysts) [25]. FY2028 consensus: not found. EBITDA consensus: not found.

  • Because consensus EPS is negative, impacts are shown as EBITDA per share. They are compared with an estimated FY2027 EBITDA base of about $5.80 per share: guide midpoint $475M × management's unofficial ~40% gross margin = $190M, less about $32M a year of cash SG&A ($2.7M a month [13]), ≈ $158M ÷ 27.5M shares.

  • This base is an estimate, not consensus.

  • Batam phase 2 (875 servers) financed and billing in 2028 — 15% chance by end-2028

    • Impact on revenue: about +$228M in FY2028. Not comparable to consensus, since no FY2028 consensus was found.

    • Impact on EPS: about +$5.80 EBITDA per share in FY2028, before any dilution from financing. That is about +100% vs the FY2027 base estimate.

    • How we got the number (estimate): 300 servers ≈ $75–80M a year [13] → about $0.26M per server per year × 875 = $228M × ~70% EBITDA margin ÷ 27.5M shares ≈ +$5.80.

    • Expected value: 0.15 × $5.80 ≈ +$0.87 per share (EBITDA).

    • What it is: the 875-server phase 2 of the $2.5B Batam contract gets dates, power at BTM-2 and closed debt.

    • Why it's not priced in: management excludes it from guidance [13], and even the above-guide consensus cannot cover it within FY2027.

    • Trigger to watch: a closed Batam project-finance facility, a BTM-2 power contract, and the naming of the NeutraDC buyer [36].

    • Why this probability: phase 1 itself is not yet financed or accepted, and BTM-2 is still "being prepared" [22]. Two more gates must open after phase 1, so it is well below one in five.

  • The remaining 700 Batam phase-1 servers go live in H2 2027 — 35% chance by 2027-12-31

    • Impact on revenue: about +$90M gross in FY2027. Only about half is unpriced, since consensus already sits about $100M above guidance. So about +$45M, or +8% vs consensus FY2027.

    • Impact on EPS: about +$1.15 EBITDA per share, about +20% vs the FY2027 base estimate.

    • How we got the number (estimate): 700 servers × $0.26M a year × half a year ≈ $90M; half unpriced ≈ $45M × 70% EBITDA margin ÷ 27.5M ≈ +$1.15.

    • Expected value: 0.35 × $1.15 ≈ +$0.40 per share (EBITDA).

    • What it is: the rest of phase 1 (tranches originally planned for December 2026 and H1 2027 [21]) is delivered and accepted in time to bill for part of 2027.

    • Why it's not priced in: management excludes it from the $450–500M guide [13].

    • Trigger to watch: acceptance of the first 300 servers (due late November or early December) and the closing of the 70% debt.

    • Why this probability: the contract and landlord capacity exist [22], but every Gorilla date has slipped (Bear #4), and debt is still unclosed.

  • The US data-centre acquisition closes and is leased up — 35% chance of closing by 2026-12-31

    • Impact on revenue: not quantifiable. The rent roll, lease terms and price are undisclosed [23].

    • Impact on EPS: not quantifiable, for the same reason. It could also be negative if it is equity-funded.

    • How we got the number (estimate): n/a.

    • Expected value: not quantifiable.

    • What it is: an operating 6 MW site with a path to 36 MW, the first owned data centre outside Asia.

    • Why it's not priced in: it is a non-binding LOI, and the stock fell 6.8% on the news.

    • Trigger to watch: a definitive agreement and its financing source.

    • Why this probability: it is non-binding with due diligence open, a late-November target, and Gorilla's record of slipping dates.

  • The Yotta IPO completes and de-risks the main counterparty — 50% chance by 2027-06-30

    • Impact on revenue: none directly. It protects about $350M of FY2027 guided revenue rather than adding to it.

    • Impact on EPS: not quantifiable. It lowers the probability of the Yotta headwind below instead.

    • How we got the number (estimate): n/a.

    • Expected value: not quantifiable.

    • What it is: Yotta files its DRHP, raises up to $1.5B and repays debt [35].

    • Why it's not priced in: the stock trades as if Yotta revenue is heavily discounted.

    • Trigger to watch: a DRHP on SEBI's site, and whether it discloses the Gorilla contracts.

    • Why this probability: the "within 2 weeks" (July) promise is already more than three months late [34]. Indian IPO windows are open, and pre-IPO money is raised [35].

Headwinds NOT in the forecasts

  • Delivery or acceptance slips push ≥20% of FY2027 revenue into 2028 — 45% chance in FY2027

    • Impact on revenue: −$115M, −20% vs consensus FY2027 ($574.9M).

    • Impact on EPS: about −$3.13 EBITDA per share, about −54% vs the FY2027 base estimate.

    • How we got the number (estimate): $115M of lost GPUaaS revenue × 75% gross margin (costs mostly depreciation, kept in EBITDA as lost margin) ≈ $86M ÷ 27.5M ≈ −$3.13.

    • Expected value: 0.45 × −$3.13 ≈ −$1.41 per share (EBITDA).

    • What it is: a one-quarter slip on about $425M of annualised Yotta and Batam revenue already costs about $106M.

    • Why it's not priced in: consensus sits about $100M above the top of the guide [25].

    • Trigger to watch: Q3 results (about 2026-11-18, unconfirmed): any Yotta 1 billing in Q4 guidance, and Batam acceptance.

    • Why this probability: three programmes have each slipped at least once (Bear #4), but the hardware is now moving, so the slips are shrinking.

  • The March 2027 reset re-prices the notes near today's price — 85% chance on 2027-03-15

    • Impact on revenue: none.

    • Impact on EPS: about −28% per share on a fully diluted basis, ≈ −$1.62 EBITDA per share vs the base estimate.

    • How we got the number (estimate): fully diluted shares 27.5M + 9.1M = 36.6M now → 27.5M + 23.0M = 50.5M at a $10.10 reset; 36.6 ÷ 50.5 = 0.72 → −28% × $5.80 ≈ −$1.62.

    • Expected value: 0.85 × −$1.62 ≈ −$1.38 per share (EBITDA).

    • What it is: the 5-day average price before 2027-03-15 sets a lower conversion price for both series [6,7].

    • Why it's not priced in: broker targets ($24–44) and consensus EPS use about 27.5M shares [25].

    • Trigger to watch: the share price in early March 2027, and any equity raise of ≥75% of principal, which would end the resets.

    • Why this probability: the stock must more than double, from $10.10 to above $25.48, in five months to avoid it. Only clear billing proof could do that.

  • GPUaaS gross margin comes in near 50%, not 75% — 35% chance in FY2027

    • Impact on revenue: none.

    • Impact on EPS: about −$3.86 EBITDA per share, about −67% vs the base estimate.

    • How we got the number (estimate): about $425M of FY2027 GPUaaS revenue × 25 margin points ≈ $106M ÷ 27.5M ≈ −$3.86.

    • Expected value: 0.35 × −$3.86 ≈ −$1.35 per share (EBITDA).

    • What it is: power, connectivity, memory inflation and landlord fees eat the spread on fixed-fee contracts.

    • Why it's not priced in: the bullish models (e.g. Compass Point's FY2027 EBITDA of $772.7M on $1.2B revenue [28]) assume about 75% rental margins.

    • Trigger to watch: the first quarter with rental revenue, and gross margin on that line.

    • Why this probability: the cost structure is undisclosed, and memory costs rose 40–60% while hardware price locks run only to December [13]. But fixed fees and possible pass-through clauses limit the damage.

  • Yotta renegotiates, delays payment or partly defaults — 15% chance within 36 months

    • Impact on revenue: up to −$350M, −61% vs consensus FY2027 in a full stop. That is the extreme case.

    • Impact on EPS: up to −$8.90 EBITDA per share, which would wipe out the base estimate.

    • How we got the number (estimate): $350M × ~70% EBITDA margin ÷ 27.5M ≈ −$8.90.

    • Expected value: 0.15 × −$8.90 ≈ −$1.34 per share (EBITDA).

    • What it is: an IPO failure or funding gap at Yotta leads to a slower ramp or renegotiated fees.

    • Why it's not priced in: guidance and consensus treat Yotta revenue as contracted [13].

    • Trigger to watch: no DRHP by December, Indian press on Yotta funding, and rising Indian receivables at Gorilla.

    • Why this probability: Nvidia's offtake and the pre-IPO raise support Yotta, but its own IPO timetable keeps slipping [34,35].

  • New equity or equity-linked raise within 12 months — 60% chance by 2027-10-12

    • Impact on revenue: none.

    • Impact on EPS: about −29% per share for a $100M raise at about $9, ≈ −$1.65 EBITDA per share.

    • How we got the number (estimate): $100M ÷ $9 ≈ 11.1M new shares; 27.5 ÷ 38.6 = 0.71 → −29% × $5.80 ≈ −$1.65.

    • Expected value: 0.60 × −$1.65 ≈ −$0.99 per share (EBITDA).

    • What it is: equity to fund Batam and Yotta 2 equity portions, the US acquisition, or Korat.

    • Why it's not priced in: the targets use the current share count.

    • Trigger to watch: a new F-3 or 6-K. Note that the July SPA standstill (no new shares until 30 days after the resale registration became effective) has likely lapsed [8].

    • Why this probability: Gorilla raised equity or converts in July 2025, June 2026 and July 2026 [33]. It has $228M of commitments and unclosed debt. The 73.5M authorised-share ceiling may force a shareholder vote first [9].

  • A further Egyptian pound devaluation of 10% on the remaining contract — 30% chance within 12 months

    • Impact on revenue: −$7.3M, −1.3% vs consensus FY2027 (spread over 2026–29).

    • Impact on EPS: −$0.26 EBITDA per share, −4.5% vs the base estimate.

    • How we got the number (estimate): $72.7M remaining [1] × 10% ≈ $7.3M (costs mostly in USD, so it falls straight to EBITDA) ÷ 27.5M ≈ −$0.26.

    • Expected value: 0.30 × −$0.26 ≈ −$0.08 per share (EBITDA).

    • What it is: further EGP weakness, with the IMF programme ending.

    • Why it's not priced in: consensus is in USD at current rates.

    • Trigger to watch: the IMF 8th review (mid-December) and the EGP rate.

    • Why this probability: the Reuters poll expects about 49 per dollar by June 2027 (stable), but the end of the programme raises the risk of a step move [44].

Note on ranking: the top four headwinds have almost the same expected values (−$1.34 to −$1.41). They are not independent: a delivery slip lowers the share price, which deepens the reset and raises the odds of a new raise.

Investment Risk Score: 72/100 (high)

# Factor Weight Measured value Score Why
1 Liquidity & funding runway 15% Cash $179.4M (06-30) + about $120M net July notes ≈ $300M (estimate); PP&E commitments $228.0M; LTM operating cash flow ≈ −$20.7M; H1 capex $14.1M and "will rise substantially"; undrawn bank lines $3.9M; no project debt closed; no going-concern language 68 Cash covers commitments with about $70M left, which gives 12–36 months on overhead and interest. But the 70% project debt is unclosed, so any new tranche needs new money.
2 Leverage 10% Debt principal ≈ $245M ($232M notes + about $13M bank); adjusted EBITDA H1 −$14.6M (LTM negative); net cash ≈ $55M before commitments, net debt ≈ $175M after 80 "Large debt with negative EBITDA". It is kept just below the top band only because cash still exceeds gross debt today.
3 Dilution 10% Shares 26.19M → 27.51M (+5.0% in 7 months; +52% since 2025-01-01); notes 9.1M shares now, 23.0M at a $10.10 reset, 33.5M at the floors (122%); warrants 9.58M; CVRs 4.51M; authorised 73.5M 88 The overhang is above +25% under any reset outcome, and the reset dates are fixed.
4 Profitability 10% Quarterly gross margin 37% → 30% → 21% → −4%; H1 IFRS operating margin −60%; adjusted EBITDA −$14.6M; H1 operating cash flow −5.5% of revenue 75 Clearly loss-making, and getting worse quarter by quarter, not better.
5 Revenue quality & concentration 10% Egypt 76.5% of FY2025; three customers >10% each in H1 2026; 67% of H1 revenue one-off hardware; IFRS backlog $72.7M; Yotta ≈ $350M of $475M FY2027 guide midpoint (≈74%) 80 Over 35% from one counterparty, both in the past (Egypt) and in the plan (Yotta).
6 Execution & track record 10% Revenue guidance met or raised 4 of 4 times since FY2025; EBITDA, cash-flow and gross-margin promises missed (FY2025); 3 programmes slipped 1–3 months; "project financing next" (May) undelivered; interim statements unaudited and unreviewed after an auditor swap 74 Frequent misses on everything except the revenue line. No restatement or disclosed material weakness keeps it out of the top band.
7 Competitive position & disruption 10% No owned data centre (PP&E $29.4M); own software $3.5M of FY2025 revenue; Firmus 170k GPUs at Batam from Q1 2027; E2E 75% EBITDA margin; IndiaAI ₹65/hr 66 Easy-to-copy middleman. Protected only by contract terms that are not public.
8 Cyclicality & commodity/macro sensitivity 10% ~100% tied to AI-compute spending; B200 forward −10% vs spot; memory costs +40–60%; 2024–25 H100 rental price −60% or more 65 Highly cyclical end-market. Five-year fixed fees dampen it if customers stay solvent.
9 Regulatory, legal & geopolitical 5% US export controls (BIS China-HQ guidance), Supermicro co-founder indictment, Egypt FX and IMF, Taiwan base, Indonesia data-centre power rules pending; SPAC earn-out suit (no amount disclosed) 62 Several material exposures, none existential on its own.
10 Valuation & expectations 10% Market value ≈ $278M ≈ 0.48x FY2027 consensus revenue; consensus above guide; short interest 17.6% of shares, 10.6 days to cover; price near 52-week low 58 Cheap on stated numbers, but with heavy shorting and a story that has to deliver within five months.

Overall: weighted average = 0.15×68 + 0.10×80 + 0.10×88 + 0.10×75 + 0.10×80 + 0.10×74 + 0.10×66 + 0.10×65 + 0.05×62 + 0.10×58 = 71.9 → 72.

Change vs last: −4 vs 76 (old 6-area method). The risk level is not lower. The v3 table credits Gorilla with no near-term maturities, no covenants and about $300M of estimated cash. The old method folded those into "financial health". The table now scores the reset-driven dilution (88) and leverage (80) explicitly.

Ranking check: GRRR is now the riskiest holding in the ledger, above IPX (67) and DUOT (66). It has DUOT's single-counterparty GPU-landlord profile, plus convertible debt equal to about 84% of its market value with fixed reset dates. Unlike pre-revenue IPX, it has revenue, but that revenue currently earns a negative gross margin.

Conclusion

  • What you are really betting on: that Yotta 1 and the first Yotta 2 and Batam tranches are accepted and billing at high margins soon enough, and visibly enough, to pull the share price well above $10 before the 2027-03-15 reset. If that happens, the GPU contracts become per-share value. If not, much of that value moves to noteholders and new equity.

  • The 2–3 things that will decide the outcome:

    1. The Q3 report (about 2026-11-18) and Q4 guidance. Is there any rental billing, and at what margin?
    2. Does project debt close for Batam and Yotta 2 without new equity?
    3. The share price going into 2027-03-15, because it sets the reset.

  • Signposts:

    • Bull case confirmed by:
      • Yotta 1 acceptance announced, with Q4 revenue guided at or above $60–70M.
      • A closed Batam or Yotta 2 project-finance facility on disclosed terms.
      • Yotta's DRHP filed and naming the Gorilla contracts.
      • Gross margin back above 25% in Q4 2026 or Q1 2027.
    • Bull case broken by:
      • A new equity or equity-linked raise before rental billing starts.
      • The Batam first-300 acceptance slipping into 2027.
      • Credit-loss allowance or overdue receivables rising again in the Q3 accounts.
      • A FY2027 guide cut, or the 5-day price before 2027-03-15 below about $10.

  • New since the last report:

    • The convertible indentures were read. There are six reset dates starting 2027-03-15, the June floor is $6 (verified), and up to 33.5M note shares are possible against a 73.5M authorised-share ceiling.
    • Yotta hardware is physically arriving: the first B200 phase was complete in late September, and B300 deliveries start in October. But no acceptance has been announced, and the stock has fallen from about $12 to $10.10.
    • $228M of property and equipment commitments are confirmed. The project debt promised as "next" in May is still not closed. A non-binding LOI for a US data centre adds a new capital need.

Open questions & data gaps

  • GPU ownership and revenue recognition:
    • Answered in part: the $228M commitment is to buy property and equipment, and management depreciates over 5 years. Both point to ownership.
    • Still open: the $52.5M of H1 hardware sales sit in Security Convergence at a point in time, and the buyer is undisclosed.
    • High importance: this decides whether 2026 revenue is resale or rental.
  • Net July proceeds and current cash: the about $120M net is a reported figure; there is no company cash update after 06-30. Medium importance.
  • Batam:
    • The customer identity, take-or-pay text and acceptance of the first 300 servers were not found.
    • The project-debt terms were not found.
    • High importance.
  • Backlog reconciliation: "$5B signed" against $72.7M IFRS is still unexplained, as is the Egypt revenue window moving to 2029. Medium importance.
  • Q2 call: only AI-generated or summary transcripts were available. Quotes are summary-level. Medium importance.
  • Yotta: no DRHP found, and Yotta's debt was not found in a primary source. High importance.
  • Benchmark "Strong Buy" (2026-09-29): still single-source. Low importance.
  • Q3 2026 results date: about 2026-11-18 (AJ Bell, unconfirmed). Last year's was 2025-11-17. Low importance.
  • The H1 operating-loss conflict ($6.1M IFRS-derived vs "$2.2M" in the call summary) and the F-3 vs 6-K equity figures are unexplained. Low-Medium importance.
Sources
  1. H1 2026 condensed interim consolidated financial statements (Ex. 99.2, Form 6-K) — SEC EDGAR — 2026-08-24 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926028827/ex_974013.htm
  2. H1 2026 results press release (Ex. 99.1, Form 6-K) — SEC EDGAR — 2026-08-24 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926028827/ex_974012.htm
  3. Q1 2026 results press release and statements (Ex. 99.1/99.2) — SEC EDGAR — 2026-05-27 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926018444/ex_944447.htm
  4. FY2025 results press release — SEC EDGAR — 2026-03-02 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926006382/ex_926172.htm
  5. Form 20-F FY2025 — SEC EDGAR — 2026-04-15 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926012385/grrr20241231_20f.htm
  6. Indenture, June 2026 notes (Ex. 99.2, Form 6-K) — SEC EDGAR — 2026-06-05 — https://www.sec.gov/Archives/edgar/data/1903145/000121390026065898/ea029373601ex99-2.htm
  7. Indenture, July 2026 notes (Ex. 99.2, Form 6-K) — SEC EDGAR — 2026-07-17 — https://www.sec.gov/Archives/edgar/data/1903145/000121390026079193/ea029837101ex99-2.htm
  8. Securities Purchase Agreement, July 2026 notes (Ex. 99.1) — SEC EDGAR — 2026-07-17 — https://www.sec.gov/Archives/edgar/data/1903145/000121390026079193/ea029837101ex99-1.htm
  9. Form F-3 (resale registration, July notes) — SEC EDGAR — 2026-08-12 — https://www.sec.gov/Archives/edgar/data/1903145/000121390026087968/ea0301044-f3_gorilla.htm
  10. Form 6-K (prospectus supplement notice to holders) and 6-K filing list — SEC EDGAR — 2026-10-08 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926032361/grrr20261008_6k.htm
  11. Form 6-K, change in certifying accountant (Marcum Asia → UHY) — SEC EDGAR — 2026-05-06 — https://www.sec.gov/Archives/edgar/data/1903145/000143774926015208/grrr20260506_6k.htm
  12. Q3 2025 results and FY2026 guidance (Ex. 99.1) — SEC EDGAR — 2025-11-17 — https://www.sec.gov/Archives/edgar/data/1903145/000143774925035422/ex_860578.htm
  13. Gorilla Q2 2026 earnings call transcript (AI-assisted, Motley Fool text; summary only) — Webull — 2026-08-24 — https://www.webull.com/news/15499649423107072
  14. Gorilla Q1 2026 earnings call summary — AllMind — 2026-05-27 — https://allmind.ai/earnings/event/bt_GRRR_2026_Q1
  15. GRRR: The Good, The Bad and The Ugly (issuer-sponsored) — Zacks SCR — 2026-08-25 — https://s27.q4cdn.com/906368049/files/News/2026/Zacks_SCR_Research_08252026_GRRR_Lantier.pdf
  16. Fitch assigns Gorilla first-time 'B-' rating, Outlook Positive (company release) — Newsfile — 2026-08-27 — https://www.newsfilecorp.com/release/311788/
  17. Gorilla & Yotta expand India AI infrastructure collaboration (~US$2.8B) — Yotta — 2026-04-30 — https://yotta.com/?p=13676
  18. Gorilla Technology accelerates execution across its Asian AI infrastructure platform — Gorilla IR — 2026-07-21 — https://investors.gorilla-technology.com/gorilla-technology-accelerates-execution-across-its-asian-ai-infrastructure-platform/
  19. Gorilla moves Yotta AI infrastructure programme into physical execution — Newsfile — 2026-09-08 — https://www.newsfilecorp.com/release/313333/
  20. Yotta's Nvidia GPU push enters execution phase as Gorilla deliveries begin — Communications Today — 2026-10-05 — https://www.communicationstoday.co.in/yottas-nvidia-gpu-push-enters-execution-phase-as-gorilla-deliveries-begin/
  21. Gorilla secures US$2.5B five-year AI GPUaaS compute contract (Batam) — Newsfile — 2026-06-23 — https://www.newsfilecorp.com/release/302522/
  22. Data centre fully secured before launch, Telkom accelerates NeutraDC Batam — NeutraDC — 2026-06-09 — https://www.neutradc.com/news/60643/data-center-fully-secured-before-launch-telkom-indonesia-accelerates-expansion-of-neutradc-capacity-in-batam
  23. Gorilla Technology eyes 36 MW US data center acquisition and over 100 MW expansion (non-binding LOI) — Blackridge Research / Media OutReach — 2026-10-09 — https://www.blackridgeresearch.com/news-releases/gorilla-technology-eyes-36-mw-us-data-center-acquisition-and-over-100-mw-expansion
  24. Gorilla Technology secures multi-year Taiwan AI contract worth tens of millions of dollars — Finviz / InvestorsHub — 2026-09-22 — https://finviz.com/news/394360/gorilla-technology-secures-multi-year-taiwan-ai-contract-worth-tens-of-millions-of-dollars
  25. GRRR overview, price and forecast (S&P consensus) — StockAnalysis — 2026-10-12 — https://stockanalysis.com/stocks/grrr/forecast/
  26. GRRR price target history — MarketBeat — 2026-09-29 — https://www.marketbeat.com/stocks/NASDAQ/GRRR/price-target
  27. GRRR forecast — TipRanks — 2026-10 — https://www.tipranks.com/stocks/grrr/forecast
  28. GRRR's $5.5B commercial book: Compass Point initiation — Stocktwits — 2026-08-18 — https://stocktwits.com/news-articles/markets/equity/grrr-5-5b-commercial-book-wall-street-attention-compass-point-193-upside/cZYcWj8RJj0
  29. Why is Gorilla Technology stock falling Tuesday? — Benzinga via Webull — 2026-08-25 — https://www.webull.com/news/15462915176678400
  30. GRRR short interest (FINRA data) — Equibles — 2026-10-12 — https://equibles.com/stocks/grrr/shortinterest
  31. Gorilla sinks 9% as data center pivot keeps selling off — 24/7 Wall St. — 2026-10-12 — https://247wallst.com/investing/2026/10/12/gorilla-sinks-9-as-data-center-pivot-keeps-selling-off-bigbear-ai-dips-palantir-slips/
  32. Forms 3, 4 and 144 for Gorilla insiders — SEC EDGAR — 2026-03 to 2026-08 — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=1903145
  33. GRRR stock tracks worst rout in over a year — Stocktwits — 2026-07-16 — https://stocktwits.com/news-articles/markets/equity/grrr-tracks-worst-rout-over-year-what-investors-watching/cZZEP69R7rx
  34. Yotta Data Services to file DRHP with SEBI within 2 weeks, says MD — Informist Media — 2026-07-02 — https://informistmedia.com/EquityWire/53918/INTERVIEW-Yotta-Data-Services-to-file-DRHP-with-SEBI-within-2-wks-says-MD
  35. Indian data centre operator Yotta targets 2027 IPO, seeks up to $1.5b (Reuters interview) — DealStreetAsia — 2026-09-02 — https://www.dealstreetasia.com/stories/data-centre-operator-yotta-india-ipo-493935
  36. Divestasi NeutraDC, Telkom: evaluasi binding offer, ditarget rampung akhir 2026 — Bisnis.com — 2026-09-07 — https://teknologi.bisnis.com/read/20260907/101/2002058/divestasi-neutradc-telkom-evaluasi-binding-offer-ditarget-rampung-akhir-2026
  37. Firmus, Nvidia plan Batam AI data centre (170,000 accelerators) — TechRepublic — 2026 — https://www.techrepublic.com/article/news-apac-indonesia-firmus-nvidia-ai-data-center/
  38. E2E Networks Q1 FY27 press release — NSE — 2026-07-21 — https://nsearchives.nseindia.com/corporate/E2E_21072026135751_Press_Release21072026.pdf
  39. B200 rental price August 2026 update — Silicon Data — 2026-09 — https://www.silicondata.com/blog/b200-rental-price-august-2026-update
  40. Democratising AI in India (IndiaAI Mission explainer) — PIB, Government of India — 2026-02-10 — https://static.pib.gov.in/WriteReadData/specificdocs/documents/2026/feb/doc2026210784401.pdf
  41. Super Micro Computer Form 10-K FY2026 (and TrendForce DRAM pricing via secondary summary) — SEC EDGAR / TrendForce — 2026-08-31 — https://www.sec.gov/Archives/edgar/data/1375365/000137536526000022/smci-20260630.htm
  42. Three charged in the US with smuggling AI chips into China — Al Jazeera — 2026-03-20 — https://www.aljazeera.com/news/2026/3/20/three-charged-in-the-us-with-smuggling-ai-chips-into-china
  43. US takes step to halt Nvidia AI chip shipments to Chinese firms outside China (Reuters) — Khaleej Times — 2026-06-01 — https://www.khaleejtimes.com/tech/us-takes-step-to-halt-nvidia-ai-chip-shipments-to-chinese-firms-outside-china-2
  44. IMF Board completes Egypt's 7th EFF review — Ahram Online — 2026-07 — https://english.ahram.org.eg/News/573797.aspx
  45. Nvidia Form 10-Q (quarter ended 2026-07-26) — SEC EDGAR — 2026-08-26 — https://www.sec.gov/Archives/edgar/data/1045810/000104581026000075/nvda-20260726.htm
  46. Problems at Gorilla Technology Group — The Bear Cave — 2025-03-06 — https://thebearcave.substack.com/p/problems-at-gorilla-technology-group
  47. CoreWeave Form 10-Q (June 2026 quarter) — SEC EDGAR — 2026-08-12 — https://www.sec.gov/Archives/edgar/data/1769628/000176962826000366/crwv-20260630.htm
  48. Gorilla touts shift to AI infrastructure in Water Tower Research fireside chat — The Globe and Mail / MarketBeat — 2026-05-07 — https://www.theglobeandmail.com/investing/markets/stocks/GRRR-Q/pressreleases/1819472/gorilla-technology-group-touts-shift-to-ai-infrastructure-3-3b-yotta-gpu-framework-in-fireside-chat/
  49. GRRR stock eyes best month in over a year (CFO on "secured but not publicly disclosed" contracts) — Stocktwits — 2026-08-25 — https://stocktwits.com/news-articles/markets/equity/grrr-hidden-ai-upside-beyond-guidance-unannounced-wins/cZYk0nrRJ8a
  50. Gorilla Technology secures $2B Supermicro supply deal — LetsDataScience — 2026-06-02 — https://letsdatascience.com/news/gorilla-technology-secures-2b-supermicro-supply-deal-f283be8f
  51. Yotta and Nvidia DGX Cloud Lepton / Shakti Cloud — Yotta — 2025 — https://yotta.com/?p=10449
All GRRR Bulls vs Bears reports
Disclaimer. AI-generated research and podcast (Claude by Anthropic, voices by Google Gemini) for educational purposes — not financial advice and not a recommendation to buy or sell. Willem (MR.INVEST) owns this stock. Check the company’s own filings before making any decision.
← All Bulls vs Bears reports
GRRR Bulls vs Bears — Stock Analysis — MR.INVEST