Author: a16z crypto; Translator: Felix, PANews
1. Prediction Markets Are Larger, More Comprehensive, and More Intelligent
—Andy Hall, Cryptocurrency Research Advisor at a16z, Professor of Political Economy at Stanford University
Prediction markets have become mainstream, and by 2026, with their integration with cryptocurrencies and AI, their scale, coverage, and intelligence will only increase further, while also bringing new and significant challenges to their builders.
First, more contracts will be listed this year. This means that not only will real-time odds be available for major elections or geopolitical events, but also real-time odds for various detailed outcomes and complex, intertwined events.
As these new contracts disclose more information and integrate into the news (which is already happening), they will raise important social questions, such as how to balance the value of this information and how to better design them to be more transparent and auditable—and cryptocurrencies can precisely address this. To handle the sheer volume of contracts, new methods are needed to reach consensus and resolve issues within the contracts. While centralized platforms are important for determining whether events actually occurred (how to confirm this), their limitations have been highlighted in controversial cases like the Zelensky "suit incident" and the Venezuelan election market. To address these extreme cases and help expand prediction markets into more practical applications, new decentralized governance and LLM oracles can help determine the truth behind controversial outcomes. AI opens up more possibilities for oracles beyond LLM. For example, AI agents trading on these platforms can search for global signals to gain short-term trading advantages, revealing new worldviews and ways of predicting future events. Beyond acting as sophisticated political analysts capable of querying insights, these agents can also reveal new information about the fundamental predictors of complex social events when their emerging strategies are studied. Will prediction markets replace polls? No; they will make polls better (and poll information can be passed on to prediction markets). As a political scientist, what excites me most is how prediction markets can work in conjunction with rich and dynamic polling systems—but it will also rely on new technologies like AI, which can improve the polling experience; and cryptography, which can provide new ways to prove that poll/survey respondents are not bots but real people, etc. 2. This year, cryptography will provide a new foundational tool for industries outside of blockchain. —Justin Thaler, member of the a16z cryptography research team, Associate Professor of Computer Science at Georgetown University. SNARKs (cryptographic proofs that verify computation without re-performing the computation) have been used primarily in the blockchain space for years. The overhead is simply too great: proving a computation can take a million times more work than directly running the computation. This might be worthwhile with thousands of validators sharing the cost, but impractical in other situations. But that's about to change. This year, the overhead of zkVM provers will drop by about 10,000 times, with a memory footprint of only a few hundred megabytes—fast enough to run on a mobile phone and inexpensive enough to run anywhere. 10,000 times is a magical number, partly because high-end GPUs have about 10,000 times the parallel throughput of a laptop CPU. By the end of 2026, a single GPU will be able to generate proofs that would otherwise be executed by a CPU in real time. This promises to realize a vision from an earlier research paper: verifiable cloud computing. If you're already running CPU workloads in the cloud—because your computational demands aren't sufficient for GPUs, or you lack the expertise, or for historical reasons—you'll be able to obtain cryptographic correctness proofs at a reasonable cost. The provers are already optimized for GPUs; your code won't need to be optimized. 3. Witnessing the Rise of “Staking Media” — Robert Hackett, a16z Cryptocurrency Editorial Team The traditional media model (and its so-called objectivity) has long shown cracks. The internet has given everyone the right to speak, and more and more operators, practitioners, and builders are beginning to engage directly with the public. Their views reflect their interests in the real world, and surprisingly, audiences often respect them not because of these interests, but because of those interests. The new change here is not the rise of social media, but the emergence of crypto tools that allow people to make publicly verifiable commitments. As AI makes generating unlimited content cheap and easy (claiming anything, whether the viewpoint or identity is real or fictitious), relying solely on the voices of the masses (or bots) is no longer sufficient. Tokenized assets, programmable lock-up periods, prediction markets, and on-chain historical records provide a more solid foundation for trust: commentators can express opinions and demonstrate consistency between their words and actions. Podcasters can lock up tokens to show they are not speculatively manipulating the market or "pump and dumping." Analysts can link their predictions to publicly settled markets, thus establishing an auditable track record. This is the rudimentary form of what I understand as "staking media": media that not only endorses stakeholder beliefs but also provides corresponding proof. In this model, credibility comes neither from "word of mouth" nor from unfounded assertions; instead, it comes from having stakes that allow for transparent and verifiable commitments. "Staking media" will not replace other media forms but rather complement existing ones. It provides a new signal: not just "Trust me, I am neutral," but "This is the risk I'm willing to take, and you can verify the truth of what I say."
Alex
Brian
Kikyo