Embargo. Sealed Released

Results nobody can open before release time

01 — The hour before

A result that can be read an hour early can be traded an hour early. Polls, award winners, injury reports and earnings now move public betting markets, and every name on the distribution list is a way in. Embargo removes that hour: nobody can open the file until release time, and every early copy carries its reader's mark.

Plate 01, Sealed send: a sculptural object photographed on a pale studio backdrop — Pl. 01 — Sealed send1024 × 1024

01 / 04

Upload the results and set the release time.

Each file is encrypted to the drand round for that moment. The key does not exist until then, for recipients, wire services or us.

Plate 02, Marked copies: a wide studio photograph of a sculptural object — Pl. 02 — Marked copies1344 × 768

02 / 04

Editors who must see the numbers early each get their own copy.

Small differences in wording, number formatting or image detail. A leaked version points to the copy it came from; a paraphrase from memory does not.

Plate 03, Release registry: a square material study — Pl. 03 — Release registry1024 × 1024

03 / 04

Release times are logged privately, and made public the moment they pass.

The pre-release window for each result is then on record, for the report and for anyone you show it to.

Plate 04, Tripwire report: a long horizontal studio photograph — Pl. 04 — Tripwire report1536 × 512

04 / 04

After each release, a private report on the window before it.

Polymarket and Hyperliquid positions above your threshold, opened in the pre-release window, set against the marked copies that existed. You decide whether it is published.

Embargo locks your results until release time, marks every early copy and checks the markets afterwards. It reports positions and timing. It does not name traders or stop trades.

Seal a test release

02 — Why now

On 31 October one Polymarket account bet about $200K on a Senate race roughly an hour before a major poll was published, and made about $77K. Polymarket's rules do not ban insiders, and no clear federal statute covers event contracts. The organisation that holds the result can act without waiting for a venue.

Fri 30 OctSenators Warren and Curtis write to the heads of Polymarket and Kalshi.

Sat 31 OctA Senate poll is published. A position of about $200K was opened roughly an hour before.

Sat 7 NovKalshi publishes odds on the 2028 nominations. Primary polling runs through 2027.

Wed 9 DecFOMC decision. A scheduled release that markets already trade.

Fri 11 DecFederal funding deadline. Another dated result with live markets.

Jan 2027Warren is expected to chair Senate Banking and has named prediction-market insider trading as an early priority. Not finalised.

03 — Terms

No token

Embargo is a subscription paid by the organisations that produce results. There is no token at launch, because one would add nothing a pollster or newsroom needs.

Ticker: $EMBGO is the reserved name. Nothing is issued under it.
Type: None at launch. A subscription for pollsters, newsrooms, award bodies, leagues and investor-relations teams.
Chain: Time-lock encryption uses drand tlock, off-chain. The monitor reads Polymarket trades on Polygon and Hyperliquid HIP-4 trade data.
Launch mechanic: No token. This is deliberate, not an oversight.
Holding it: There is nothing to hold, and nothing to buy.

04 — Risks

Known limits

Embargo reports positions and timing. It does not name traders or stop trades.

Inside access: Staff with legitimate early access can still trade. A marked copy identifies a leak after it happens; it does not prevent one.
Memory: Someone who paraphrases from memory carries no mark.
Venues: The monitor sees onchain venues only. Off-chain books are outside its view.
False positives: A well-researched trader can look like a leak. Reports stay private by default, and you decide what is published.
Who pays: Pollsters and newsrooms have small budgets, and some will expect the venues to pay for integrity. Unresolved.
Regulation: Venues may build their own version, or regulators may require a different approach, such as venue KYC or insider bans. Unresolved.

05 — Seal a test release

This file opens at 14:00 ET. Its key does not exist yet, so nobody can open it early, including us.

File: results_<poll>.pdf
Under embargo until: Wed 9 Dec, 14:00 ET
drand round: <round for 14:00 ET>
Key: Does not exist yet
Marked copies: 3 · <reader_a> <reader_b> <reader_c>
Tripwire threshold: <your threshold>

Sample release. There is no live app yet, and nothing on this page is sent anywhere.