Prediction markets have won mainstream attention for their ability to aggregate information and forecast real-world events with striking accuracy, but that success is now bringing heavier scrutiny. As policymakers, regulators, and market participants debate the future of the sector, Pred co-founder and CEO Amit Mahensaria argues that the industry’s long-term survival will depend on building what he calls “integrity infrastructure” — a combination of surveillance, transparent settlement, anti-manipulation safeguards, and clear public accountability.
The debate has intensified as prediction markets increasingly move from niche communities into broader public discussion. Their rise has also exposed them to familiar criticisms: that they may enable insider trading, create perverse incentives, or encourage speculative activity around events with serious ethical implications. In response, some governments have moved toward enforcement, while lawmakers in the United States have proposed restrictions on contracts tied to topics such as death and war.
Innovation Meets Oversight
The core tension is not new. Supporters of prediction markets argue that these platforms provide a unique public good by pulling dispersed information into a market price that can help interpret uncertainty. In that view, prediction markets are more than gambling venues; they are informational systems that can generate useful signals for the public, businesses, and institutions.
Critics, however, counter that informational efficiency does not erase the risks. They point to concerns about insider access, coordinated manipulation, unclear accountability, and markets built around morally fraught scenarios. The more visible the industry becomes, the harder it is for platforms to avoid these questions.
Mahensaria’s position sits between the two camps. He argues that self-regulation is essential, but incomplete. In his view, any platform that wants to be taken seriously over the long term should not wait for regulators to force reform. It should already be investing in the systems needed to protect market integrity. That includes monitoring trading activity, publishing clear settlement rules, detecting manipulation attempts, and reporting transparently to users and stakeholders.
For Mahensaria, this is not optional hygiene; it is foundational infrastructure. A platform that fails to build these protections may survive short-term hype, but it is unlikely to maintain trust through periods of controversy or volatility.
The Limits of Self-Regulation
At the same time, Mahensaria does not believe self-regulation alone is enough. He notes that many industries only embrace strong principles after a scandal forces the issue. Financial markets, aviation, and pharmaceuticals all provide historical examples of sectors that learned the limits of self-policing only after significant failures exposed structural weaknesses.
That is why he advocates proportionate regulation rather than a hands-off regime or an overly punitive one. In practical terms, this means regulators should focus on establishing baseline standards without destroying the structural advantages that make prediction markets useful in the first place.
According to Mahensaria, the areas that deserve the most regulatory attention are settlement integrity, counterparty transparency, and anti-manipulation controls. Those are the parts of the market structure that directly affect trust and fairness. By contrast, regulation that is too broad or too blunt could suppress experimentation before the technology and market design have fully matured.
This distinction matters because prediction markets are still evolving. Heavy-handed oversight at an early stage could reduce the industry’s ability to test different models, refine user protections, and discover where the technology creates real value. But a complete absence of standards could lead to scandals that undermine the entire category.
Not All Markets Carry the Same Ethical Risk
One of the article’s key themes is that prediction markets are not all alike. Mahensaria points out that platforms focused on verifiable outcomes with natural timelines may face fewer problems than markets tied to political crises, geopolitical conflict, or other highly charged events where outcomes can be subjective, manipulable, or ethically troubling.
That distinction is becoming increasingly important as lawmakers and regulators attempt to decide where the boundaries should be drawn. Markets involving assassinations, wars, or political instability raise questions that go beyond settlement mechanics. Even if such contracts can theoretically be resolved accurately, the larger issue is whether they create harmful incentives or normalize speculation on human tragedy.
Mahensaria argues that the industry should not dismiss these concerns as mere squeamishness. The real question is not only whether a market can function, but whether the information it produces is worth the moral cost of the mechanism used to produce it. That framing pushes the discussion beyond compliance and into the realm of legitimacy.
When it comes to deciding which contracts should be listed, Mahensaria supports a hybrid approach. Platforms should exercise judgment and explain their listing decisions publicly, while regulators should set boundaries for categories that are clearly harmful. In other words, curation should be both a market responsibility and a regulatory concern.
AI as a Monitoring Tool, Not an Automatic Judge
The article also highlights the industry’s growing interest in using artificial intelligence to detect suspicious trading behavior. That trend has been underscored by a recent partnership involving Polymarket, Palantir Technologies, and TWG AI to develop an AI-driven monitoring platform for sports prediction trading.
Mahensaria sees genuine promise in this approach. AI can process large datasets and identify patterns that deviate from expected behavior, especially where those patterns may correlate with insider knowledge or coordinated manipulation. In markets where timing, information asymmetry, and behavior clustering matter, that type of pattern recognition can offer a meaningful advantage.
Still, he warns against treating AI outputs as final verdicts. One of the central risks is false positives — systems that mistake legitimate skill, strong analysis, or unusual but lawful positioning for misconduct. If platforms rely too heavily on automated flags, they could end up penalizing exactly the kind of informed participation that makes prediction markets valuable.
For that reason, Mahensaria argues that AI should serve as a trigger for investigation, not an automatic punishment mechanism. Human review and contextual interpretation remain necessary. Surveillance should protect the market from abuse without turning it into a system that discourages informed traders.
He contrasts this with practices in parts of the traditional sports trading industry, where successful participants have sometimes been punished through account restrictions or reduced limits. In his view, prediction markets should avoid repeating that model. The goal is not to punish winning analysis, but to identify unfair informational advantages and manipulative conduct.
Why On-Chain Transparency Matters
Mahensaria also emphasizes that blockchain-based prediction markets may have an important structural advantage in this area. On-chain systems create transparent, immutable records of trades, providing a richer and more auditable dataset than many traditional environments. That visibility can strengthen both internal monitoring and external accountability.
When combined with AI-driven analytics, this on-chain transparency could produce a more robust integrity framework than what exists in much of conventional sports trading today. Every trade leaves a trace, and that trace can be studied for patterns of timing, coordination, and abnormal behavior. In theory, that makes it easier to investigate suspicious conduct and harder to conceal it.
But transparency alone is not enough. Data must be paired with rules, review processes, settlement clarity, and governance standards. Otherwise, even perfect records may not translate into effective oversight. The blockchain layer can improve the raw materials of integrity, but institutions and platform design still determine whether those materials are used well.
A Sector Entering a More Mature Phase
The broader takeaway from Mahensaria’s comments is that prediction markets are entering a new phase. The industry can no longer rely only on the argument that innovation should be left alone. As the category grows, it must demonstrate that it can handle ethical dilemmas, market abuse risks, and public accountability in a credible way.
That does not necessarily mean embracing maximal regulation. Instead, it suggests a more mature balance: self-regulation where platforms build serious internal controls, and external oversight where governments define baseline protections and draw lines around the most problematic categories.
The survival of prediction markets may therefore depend less on whether they can generate accurate forecasts and more on whether they can build trusted systems around those forecasts. Accuracy may have driven their rise, but integrity could determine whether they endure.