Mention markets can give an edge to people who know what someone will say or do before everyone else. They can also give an advantage to people who are able to influence the outcome. The CFTC wants exchanges to figure out who those people are.
When many of us think of insider trading, boardrooms, stock tips, and corporate secrets come to mind. But Gabriel Perez wasn’t the typical case; he wasn’t a Wall Street suit. He was the man who fed the teleprompter at the White House.
That job gave him access to something invaluable to prediction market traders: advance knowledge of what President Donald Trump would say during a speech before he uttered a single word. From December 2025 to March 2026, Perez used his unique access to trade prediction market contracts on words or phrases Trump might say. The Commodity Futures Trading Commission says he made $107,539.02 in profits from those trades.
The CFTC ordered Perez to return those profits and pay a $65,000 civil penalty. It also handed him a three-year trading ban.
Once Perez’s job and access were known, the problem was easy to see. A teleprompter operator could read the same material that would later determine whether a contract on what the president said paid out.
But just because his case was clear-cut doesn’t mean other insiders are as easy to identify. After all, lots of people could conceivably have access to insider information that would help them out on a mention market. That leaves the question of where the line should be drawn when it comes to deciding who counts as an insider and who doesn’t.
A speechwriter seems like an obvious yes, but what about an assistant, a contractor, or someone with special access to a guest list? Or what about the person who doesn’t know what will be said in advance but has influence over the person whose words or actions determine how a mention market contract settles?
That’s the question the CFTC’s Division of Market Oversight attempted to address in its September 22 staff advisory, where it “strongly encouraged” exchanges to identify “potential controllers and known insiders” connected to certain mention, attendance, and interaction contracts. The advisory focuses on the risks those connections can pose when an individual’s words or actions determine a contract’s outcome.
A Single Person Can Determine Whether a Mention Market Pays Out
Prediction market contracts on events like an election, the release of economic data, or the winner of a football game typically settle on outcomes that are outside the control of any one person. Mention markets are the exception to the rule. Unlike most event contracts, mention markets can settle based on whether someone says a specific word, attends an event, or interacts with someone.
That puts an incredible amount of power in the hands of one person or a small group of people. The CFTC says that can make the settlement condition “comparatively easier to cause, prevent, or influence for personal gain.”
Thomas Rietz, a finance professor at the University of Iowa, says having better information isn’t necessarily a problem when the goal is to make the market more accurate. The distinction, he says, is between knowing what will happen and being able to make it happen.
“If the purpose of a prediction market is to aggregate the best possible information, [insiders] (i.e., those who know private things about the outcome) may actually help in pricing accuracy,” Rietz told Gambling Insider. “However, if a trader can control the outcome, then it is more than just getting that trader’s information into the market. The trader can take advantage of others by affecting the outcome itself.”
In its staff advisory, the CFTC used the example of a contract on whether a podcast host says a “particular catchphrase.” All the host has to do to make the contract settle that way is say the phrase in question, while a trader could try to get the host to say it by submitting a question or buying an on-air acknowledgment.
Rietz went even further, giving us an example of just how far that control could go. “For a thought example, consider a market where the payoff to the contract will be $1 if someone [screams ‘fire’] in a crowded theater during a live broadcast of some event like the Academy Awards,” he said.
“A trader who is attending can buy the contract and scream [‘fire’] to receive the payoff and cause a [stampede] that injures or kills others. That’s a serious problem.”
Mention Markets Face a Higher Bar Under Existing CFTC Rules
Nothing in the CFTC’s staff advisory tells prediction markets that they can’t list mention markets. However, the CFTC’s Division of Market Oversight does say its staff may view them as “presumptively readily susceptible to manipulation.”
“The bar is certainly higher,” Noah Solowiejczyk, a partner at Fenwick and former federal prosecutor who coordinated investigations with the CFTC, SEC, and other federal regulators, told Gambling Insider. “Any exchange that lists a mention market will, however, want to be in a position to demonstrate that it has diligently met its obligations to ensure that the contract is not readily susceptible to manipulation,” Solowiejczyk said.
The advisory doesn’t create “any new binding rules or regulations.” Under current CFTC requirements, exchanges are already required to avoid listing contracts that are readily susceptible to manipulation. The views expressed in the advisory come from the division’s staff and “[do] not necessarily represent the views of the Commission.”
The division’s staff lists four areas it says are “particularly relevant” when exchanges determine whether a mention market can be listed without being too easy to manipulate. Those factors are:
- Obligations that might deter the person controlling the outcome from manipulating it
- The risk of outside pressure or influence
- “Independent verification and substantial public scrutiny”
- The strength of the exchange’s trading rules, surveillance, and other controls
That last point comes down to what the exchange itself can detect and prevent. “The exchange’s trading surveillance infrastructure is going to matter here,” Solowiejczyk told us.
The advisory says staff expect exchanges to explain those safeguards when submitting contracts for review. “Is it sufficiently sophisticated that it can be customized to ensure that each particular mention market is subject to appropriately scoped prophylactic controls?” Solowiejczyk asked. “A one-size-fits-all approach does not appear sufficient based on the DMO’s guidance.”
Traditional Insider Screening Can Miss People With Access to Outcomes
Figuring out who the “potential controllers and known insiders” are has to happen before suspicious trading ever takes place. “Under the CFTC’s notice, this starts at listing, not with the trader,” Tiffany Magri, a regulatory advisor at Smarsh, told Gambling Insider.
“In practice, that means mapping who can determine or influence the outcome, and who has advance access to scripts, prepared remarks, or guest lists,” she said. But identifying those people isn’t always straightforward. Exchanges can use public records, financial disclosures, corporate filings, and other commercially available data to identify formal connections. “The harder cases are private individuals, such as a podcast host, where those sources offer little,” Magri told us.
Part of the problem is that the systems being used to monitor insider trading were designed for a different type of relationship. “Existing surveillance was built around securities and traditional derivatives, where insiders are tied to issuers and instruments,” she said.
“Mention markets turn on individual conduct, such as a word said, an appearance made, or an interaction that occurs. Those events don’t map to issuer-based restricted lists or standard order-flow models.” Exchanges can still ask traders for employment information, examine account data, or require disclosures about personal connections. But those methods will only get them so far.
“The limit is that exchanges see only what traders disclose and what can be verified,” Magri explained. “Informal ties, such as friends or family, are far harder to detect.”
Trading Patterns Can Raise Red Flags Without Revealing Insider Connections
Even if an exchange isn’t able to identify every single personal or professional connection before trading begins, the trading itself can still provide clues. Magri told us that tools including employment checks and trading-pattern surveillance can help, “but none is sufficient alone.”
“Restricted lists work only as well as the list of insiders behind them, and they depend on someone knowing who to include,” she said. Employment checks can catch connections that are disclosed or otherwise discoverable, but Magri said they may miss someone trading through a relative or another account.
Pattern monitoring can catch something different. One example Magri gave was a new account taking an unusually large, well-timed position in a thin market, which may attract attention even if the trader doesn’t have a known connection to anyone involved in the contract.
“Pattern monitoring … can surface behavior no list would catch, but it generates false positives and shows suspicious timing, not why someone traded,” she said. The Perez case shows how that can work in practice. “Kalshi has said its surveillance flagged the activity, and the CFTC credited its assistance,” Magri said. “Someone with script access is easy for an insider list to miss, which is why trading-pattern surveillance matters.”
Spotting a suspicious trade may still be easier than identifying someone who can influence how a contract settles. “A person who controls the outcome can act within entirely normal conduct: saying a word or attending an event is not suspicious in itself,” she said.
“The trade may be the only signal, and it looks like any other well-timed bet.” This leaves exchanges trying to distinguish an ordinary well-timed trade from one based on advance knowledge or influence over the outcome.
Larger Potential Payouts Can Increase the Incentive to Manipulate Outcomes
Better screening and surveillance can help prediction markets manage insider trading risks, but Rietz said fairness also depends on how the market itself is structured. He said the problem becomes even more complicated when a single person or a small group of people can affect the outcome.
“It depends on the underlying purpose of the market, the impact of the outcome, and how the market is structured as well,” Rietz told us. “If you can buy a position where you might profit by a few hundred thousand dollars or more, it’s very tempting to influence the outcome,” he said.
“However, if you gain a hundred dollars or so and might lose your job if you’re discovered, then it’s a whole lot less tempting.” The potential reward, in other words, can change the incentive to interfere with an outcome.
Improved controls can certainly help reduce some of those risks, but they don’t answer the more basic question of why a particular market exists in the first place. “The first question I would ask is: [Is] there a real social benefit to aggregating information on a particular topic?” Rietz said. “If not, why run it? If so, then a considerable amount of thought should go into it.”
The post Who Counts as an Insider When Prediction Markets Bet on Human Behavior? appeared first on Gambling Insider.