Kalshi probes pre-announcement bets on Trump's press secretary pick

Kalshi has opened an internal probe into three wagers placed on its "next White House press secretary" market before President Trump's October 9 announcement that Katie Zacharia, a senior communications adviser at Trump Media & Technology Group, would succeed Karoline Leavitt, The Block reports, picking up the Wall Street Journal's initial reporting. Kalshi confirmed the probe to the WSJ and declined to comment further; the CFTC did not respond to The Block's request before publication.

The three bets

Per the WSJ reporting carried by The Block, Kalshi's Zacharia contract had traded at roughly a 1% implied probability in the days before the announcement. Three accounts bought in during that window:

• ~$19 at approximately 10:42 p.m. ET on Thursday, October 8 — about 15 hours before news reports began.
• ~$74 at approximately 1:41 p.m. ET on Friday, October 9 — about 20 minutes before the first wire reports.
• ~$80 at approximately 1:41 p.m. ET on Friday — same window as the second.

Expected payout at resolution: about $9,600 combined ($1,896 + $3,689 + $4,023). Whether the funds have been frozen pending Kalshi's review is unclear from the WSJ account.

Why this matters

Kalshi is a CFTC-regulated designated contract market (DCM). Event contracts settled through a DCM sit under the same anti-manipulation and insider-trading rules the CFTC applies to swaps and futures, not the lighter regime that governs social-prediction platforms. A pattern of trades placed on non-public information about a political appointment — if the trader actually had non-public information — is the exact fact pattern the Commodity Exchange Act's section 6(c)(1) covers.

The Zacharia pick was not telegraphed publicly. Kalshi traders had her at ~1%; the three accounts' positioning only pays if the market's consensus was wrong, which it was. The probe is a KYC check against those accounts to establish whether any of the three has a traceable link to the Trump Media communications shop or the White House.

Precedent

The Block flags an August 2026 CFTC settlement with a former Trump teleprompter operator over trades on Kalshi "mention-market" contracts tied to debate-script content as the closest precedent. The Zacharia probe is structurally similar: political-adjacent positioning in a thin market before a public announcement.

The broader context is Kalshi's recent regulatory arc. Political event contracts returned to its board only after the DC Circuit lifted the CFTC's attempt to block them in 2024, and the agency's October 2026 rulemaking — see our coverage of the CFTC event-contract IFR and NPRM — kept them on-exchange while carving casino-style gambling out of the swap definition. The Zacharia case is the first high-profile insider-trading question on the political book since that rulemaking.

What to watch

1. Account identity. Kalshi retains KYC records; whether the agency subpoenas them is the deciding step.
2. CFTC public filing. No order has posted on the CFTC's enforcement press page. An administrative order similar to the Perez matter is the realistic enforcement path.
3. Market-structure response. Kalshi has not published its own thresholds for suspending political-event contracts near announcements; the Perez settlement implied changes to pre-match surveillance that have not been detailed publicly.
4. The $9,600. The posted payouts have not been reported as paid out or clawed back; Kalshi's rulebook gives it discretion to void trades where manipulation is established.

Context

Kalshi's political book is small by volume compared to Polymarket's, but it is the only CFTC-regulated venue carrying contracts of this type. The site has previously resolved US presidential and congressional markets without an insider-trading enforcement event; Zacharia is a departure. The question for the CFTC and for Kalshi's compliance team is not whether the trades resolve profitably — they did — but whether the three accounts can be tied to a person with lawful duty to keep the announcement non-public.