Prediction Market Regulation Us

SKILL: US Prediction Market Regulation

Version: 1.0.0 | Domain: CFTC, Gambling Law, Event Contracts

Primary Regulator: CFTC

Jurisdiction Basis

• Event contracts (binary options on future events) = "swaps" under the Commodity Exchange Act (CEA)
• Dodd-Frank Act (2010) explicitly granted CFTC jurisdiction over prediction markets
• Platform facilitating trading of event contracts must register as:
  • Designated Contract Market (DCM) — full exchange (Kalshi model)
  • Swap Execution Facility (SEF)
  • Or operate under CFTC exemption/no-action letter

Key Statute

• CEA §5c(c)(5)(C) — "contrary to public interest" provision for event contracts
• 7 U.S.C. § 1 et seq. — Commodity Exchange Act base authority

Critical Precedents

Kalshi v. CFTC (2023-2024) — LANDMARK

• What happened: Kalshi sued CFTC after CFTC rejected their congressional election contracts
• Outcome: Court ruled in Kalshi's favor — CFTC CANNOT prohibit election event contracts under its "contrary to public interest" authority
• Impact: Opened the door for political prediction markets in the US under DCM registration
• Docket: ForecastEx LLC v. CFTC, No. 23-cv-3112 (D.D.C.)

Polymarket Consent Order (2022)

• What happened: Polymarket paid $1.4M fine for operating unregistered trading facility for US persons
• Outcome: Polymarket moved offshore (Cayman), geo-blocked US users
• Reality: US users still access via VPN; platform doesn't use KYC to enforce geo-block
• Citation: CFTC Docket No. 22-09 (2022)

PredictIt CFTC No-Action Letter 14-130 (2014)

• What it allowed: Academic prediction market with: 5,000 trader max per market, $850 max position
• 2023: CFTC revoked the letter; PredictIt sued, got temporary stay
• Iowa connection: Operated for Victoria University of Wellington; Iowa Electronic Markets is the academic blueprint
• Full text: https://www.cftc.gov/sites/default/files/idc/groups/public/@lrlettergeneral/documents/letter/14-130.pdf

CFTC v. Ooki DAO (2022) — CRITICAL FOR DEFI

• What happened: CFTC pursued enforcement against a DAO with no legal entity
• Outcome: CFTC won — decentralization does NOT exempt you from CFTC enforcement
• Impact: Smart contracts executing event contracts = operating an unregistered trading facility
• Docket: CFTC v. Ooki DAO, 22-cv-5416 (N.D. Cal.)

Intrade (2013)

• Shut down after CFTC enforcement for offering off-exchange binary options to US persons
• Enforcement action: CFTC v. Trade Exchange Network, 12-cv-1902 (D.D.C.)

Legal Operating Models

The Kalshi Model (Full Regulation)

• Register as DCM with the CFTC
• Cost: $1-5M in legal fees, 12-18 months to approval
• Ongoing: Compliance team, surveillance, reporting, margin requirements
• Benefit: Legally clear, can serve all US customers
• Limitation: CFTC can still reject specific contract types (though courts limited this post-Kalshi ruling)

The Polymarket Model (Offshore)

• Incorporate offshore (Cayman entity)
• Geo-block US users (IP blocking + TOS prohibition)
• Do NOT register with CFTC
• Risk: CFTC CAN still pursue enforcement (proved by $1.4M consent order)
• Reality: Largest prediction market in the world operates under this model; the fine was a cost of doing business

"Not a Prediction Market" Structures

Structure 1: Skill-Based Competition (DFS Analog)

• Outcome determined by skill → NOT gambling in most states
• Precedent: DraftKings, FanDuel — "skill-based contests"
• For AI: agent capability determines outcome, not random chance
• Risky states: Arizona, Louisiana, Montana, Washington (classified as gambling regardless)

Structure 2: Research/Academic Tool

• Frame as calibrating AI model accuracy, not betting on events
• Iowa Electronic Markets precedent: CFTC no-action letter for academic research
• Critical requirement: If real money + payout based on future event → regulators pierce the framing

Structure 3: Free-to-Play (Zero Legal Risk)

• No consideration = no gambling under any state law
• Revenue: advertising, data licensing, premium analytics
• Users compete for leaderboard rankings or platform-funded prizes
• Lowest risk, lowest revenue ceiling; viable as Phase 1 launch

State-by-State Risk

~40 States: Legal with compliance

Prohibited/Restricted (~10 states):

• Washington: Class C FELONY for all online gambling including skill games — ALWAYS GEO-BLOCK
• Arizona: Requires specific licensing
• Louisiana: Prohibited outside licensed venues
• Montana: Limited
• Iowa: Requires licensing (important for Nick's home jurisdiction)
• Connecticut: Requires licensing
• Tennessee: Sports betting legal; skill games gray

Required Compliance Regardless of State:

• Age verification (18+ minimum, 21+ in some jurisdictions)
• Responsible gaming disclosures
• Geo-blocking for prohibited states

Post-Loper Bright Implication

Loper Bright v. Raimondo, 603 U.S. ___ (2024) — Chevron deference overruled

• CFTC regulatory guidance on prediction markets is now persuasive, not binding
• Courts must independently interpret the CEA — aggressive structures are more contestable
• This cuts BOTH ways: CFTC can't rely on deference for expansive interpretation of "swap," but neither can platforms rely on informal agency tolerance

Iowa-Specific Notes

• Iowa Electronic Markets (IEM): University of Iowa has operated prediction markets under CFTC no-action letters since 1988 — the blueprint for academic exemption
• Iowa Code Ch. 725 (Gambling): Requires "consideration," "chance," and "prize" — skill games may be carved out
• Iowa requires licensing for real-money skill competitions — consult Iowa AG guidance before launching in Iowa
• Nick is Iowa-based: Iowa AG may be first regulator to take interest in any Iowa-operated prediction platform

This is legal research and intelligence, not legal advice. Consult qualified legal counsel before taking action.