Contrarian Corpus
short seller full deck initial thesis
2022-05-03 · 183 pages

IonQ Inc. IONQ

IonQ's 32-qubit 'world's most powerful quantum computer' is a Nikola-style hoax; its only real machine can't reliably add 1+1, and revenue stems from sham related-party deals.

Thesis

Scorpion Capital argues IonQ, the 2021 SPAC marketed as the first pure-play quantum computing company, is a fraud reminiscent of Nikola and Theranos. Based on 25 interviews with ex-employees, leading quantum experts, and purported customers, the report alleges IonQ's flagship '32-qubit world's most powerful quantum computer' does not exist — the sleek black-box photos on its website are staged, the actual device is a garage-sized skunkworks contraption, and its only operational machine is a primitive 11-qubit 'toy' that cannot reliably add 1+1 (correct only 59-70% of the time across 5,000 hired-expert trials). Revenue and bookings come from a 'sham' $20M related-party deal with a non-operational University of Maryland 'Q-Lab,' plus quid-pro-quo arrangements where IonQ allegedly paid customers like GE. CEO Peter Chapman also appears to misrepresent MIT credentials while two co-founding professors run the company as a part-time academic side-hustle.

SCQA

Situation

IonQ listed via SPAC in October 2021 as 'the first publicly traded pure-play quantum computing company,' reaching $36/share on hype around a purported 32-qubit machine touted as the world's most powerful quantum computer.

Complication

Ex-employees and quantum experts indicate the 32-qubit machine is a staged hoax; the only real device is an 11-qubit toy that mis-adds 1+1, while revenue stems from sham related-party deals and paid-for-partnership announcements.

Resolution

Scorpion is publicly short IonQ and calls for investors, regulators, and the press to scrutinize the hardware claims, related-party revenue, and the CEO's misrepresented MIT credentials before the charade unravels.

Reward

No explicit price target, but the implied downside from $8 and $1.6B market cap is toward zero — an academic research project with no real product, no IP, and fabricated revenue cannot justify a billion-dollar valuation.

The three reasons

  1. 1

    The 32-qubit 'world's most powerful quantum computer' is a Nikola-style staged hoax

  2. 2

    Revenue is driven by a sham $20M related-party deal with a non-operational UMD Q-Lab

  3. 3

    CEO Peter Chapman appears to fabricate MIT credentials; company is a professor side-hustle

Primary demands

  • Public disclosure that the 32-qubit 'world's most powerful quantum computer' does not exist as represented
  • Regulatory/investor scrutiny of related-party revenue recognition (UMD National Quantum Lab, GE, Hyundai)
  • Independent verification of CEO Peter Chapman's claimed MIT educational credentials
  • Revaluation of the stock to reflect an academic research project rather than a $1.6B commercial company

KPIs cited

1+1 accuracy on AWS-accessible IonQ machine
Correct answer (2) returned only 59-70% of the time across 5,000 hired-expert trials
System uptime
~53% per a paper by IonQ staff that the company allegedly buried
Qubit count on deployed system
11 qubits (vs. claimed 32); experts say 1,000-100,000 needed for a useful machine
Quantum volume (claimed)
'>4,000,000' per IonQ Oct 2020 press release — dismissed by experts as 'contrived' benchmark
Computers 'In Service'
Only 3 per IonQ investor deck, implying >$500M of market cap per machine
Market cap / share price
$1.6B market cap, $8/share (down 80% from $36 post-SPAC peak)
Short interest
7MM shares on 6.4MM ADV as of 5/2/22
UMD National Quantum Lab bookings
$14-20MM purported compute-time purchase from non-operational lab tripled bookings overnight
Speed vs. competing approaches
'1,000x slower' per longtime-friend physicist; a protein simulation could take ~3 years
Job queue latency on AWS Braket
30-60 minute waits, 'sometimes wait until the next day for the job to come back'

Pattern membership

Where this document fits across the library's 12 rhetorical / structural patterns. Orange cells are present in this deck; neutral cells are not.

Precedents cited

  • Nikola (staged prototype fraud)
  • Theranos (manual processing behind automated claims)
  • Allakos (ALLK) — prior Scorpion short, down 97%
  • Quantumscape (QS) — prior Scorpion short, down 61%

Composition what's on the 183 slides

Visual + textual elements counted across every slide in this deck. Hover a box for what that element is; click to see every slide in the corpus that uses it.

Slide gallery ·

All 183
No slide inventory yet

Pass-2 extraction may still be in progress for this deck.

Notes

Classic Scorpion Capital 183-page investigative short report — very text-heavy, long expert-quote transcripts with yellow highlighter, red arrows/circles on screenshots. Built on 25 consulting-network interviews (7 ex-employees, 11 quantum experts, 5 purported customers). Memorable 'can't add 1+1' framing device (p.111-115) where they hired experts to run 5,000 test adds on AWS Braket and tabulated error distribution — a rare specimen of a short-seller staging their own empirical experiment as a rhetorical device. Explicitly analogizes to Nikola (staged prototype photos) and Theranos (hidden manual processing behind automated claims). No formal valuation framework — pure fraud-exposure thesis. Stake disclosed qualitatively ('Scorpion Capital LLC is short IonQ') but not as a percentage.