Varun Datta, venture capitalist and CEO of Truth Ventures, explores why the next generation of Web3 winners will be those building platforms that enable wider innovation, rather than standalone products.
In May this year, Hyperliquid activated HIP-4 on mainnet, providing the world with fully collateralised binary outcome contracts settling to either zero or one USDH. Because each contract is fully collateralised rather than margined, there is no liquidation engine and no funding rate to manage. The position is prepaid, and the protocol’s risk surface collapses to a single question: did the oracle report the right result? The response to this move was immediate, with market participants placing more than 6 million contracts on the first day. Three weeks later, the platform listed its first U.S. macroeconomic event: the May year-over-year CPI print, which settled in June against official Bureau of Labour Statistics data.
Commentators responded to the news by suggesting that Hyperliquid had launched a predictor market in a similar guise to those offered by Kalshi and Polymarket. But according to deeper insights from Varan Datta, founder and CEO of Truth Ventures, the story is different.
“Most commentators see Hyperliquid’s activation of HIP-4 on mainnet as a product launch, but in reality, what really went out was a set of deployment rights. Hyperliquid wasn’t building a prediction market, but rather the infrastructure and machinery that lets other people build such markets. That’s a completely different business model and sits higher up the value chain, in my opinion.”
Datta took a similar view when writing CCN’s guide to the upgrade. He argued that the future of digital finance will be centred around scalable, programmable infrastructure, and that HIP-4 signals will power developer-led ecosystems.
Shipping a product and shipping a primitive
HIP-4 is the latest in a lineage of Hyperliquid standards, and the progression matters, as each one moved a piece of market creation from the core team to the permissionless edge.
- HIP-1 was about establishing native token standards for spot trading.
- HIP-2 provided bootstrapped on-chain liquidity for those tokens.
- HIP-3 opened perpetual futures to permissionless deployment.
Builders staking more than 500,000 HYPE could launch a perp market for any asset they wanted with a valid oracle feed
Datta is most interested in the results of HIP-3 development. Builder-deployed perpetual markets have now shot past the $1 billion mark with platform TradeXYZ playing a major role.
On this, Datta said the following: “Hyperliquid’s platform still gets the trading fees, the users, and the staking demand for its own tokens without having paid for the product or even the idea. Most normal companies can only ship what their product team, their employees, develop and output is capped by the number of people they hire. But Hyperliquid’s platform isn’t capped because it simply provides a substrate that other players build on top of. This puts it at a distinct competitive advantage.”
Builders are a distribution channel
To launch an outcome market, builders now have to stake 1 million Hyperliquid tokens as a deposit. The distinction that matters for anyone modelling this is that the stake is a bond, not revenue. It is not spent; it sits at risk, and validators can slash it if the builder feeds a false result into the market or otherwise breaks protocol rules.
The cost also decreases with scale. Since builders write the rule book for each market, the single-state slots support a rolling series of markets to amortise the stake over time. If a prediction market follows CPI prints every month, the cost of following one print is high, but the cost of 1,000 is trivial.
Datta sees this fact as more significant than a determination of trading mechanics.
“The stake Hyperliquid demands is not a fee, but rather a compliance mechanic. Builders need to underwrite honesty into their platforms using their own tokens, and the economics of remaining honest improve as they keep deploying. This is different from the conventional practised listing model used by PolyMarket and Kalshi. In those models, platforms decide which markets exist. But with Hyperliquid, the staking infrastructure itself determines which builders are allowed to run markets.”
The distinction is between an editorial gate and an economic one. A listings committee scales with headcount and is subject to jurisdictional pressure; a bond scales with capital and is enforced by slashing. The second is the more censorship-resistant design, but it also means market quality is only ever as good as the size of the bond relative to the profit available from a dishonest settlement, which is the number worth watching as deployed volume grows.
Shared books and the complementarity trick
Hyperliquids’ architecture reinforces this process on a secondary level. “Yes” and “no” orders are merged onto a shared book across builders, reducing the challenges that many platforms face with thin markets. Buying on the “yes” side at a given price is mathematically equivalent to selling on the “no” side at its complement, reducing the need for smaller operators to bootstrap the two thin books.
Currently, Hyperliquid processes over $219 billion in volume and has more than 1.4 million active traders. This means it’s able to launch new markets hot.
“This is the compounding asset that we’re so excited about at Truth Ventures,” Varun Datta explains. “Builders get liquidity for execution from day one, so Hyperliquid is effectively offering a shortcut to platform maturity. It’s the same reason why most firms now use AWS for their servers instead of constructing them in-house. It’s just easier to do that.
What Permissionless Deployment Does to HYPE
The token-economic read is the part most product-framed coverage misses.
- Every deployment locks supply. HIP-3 and HIP-4 bonds are capital withdrawn from circulation for as long as a builder wants to operate, so builder count translates fairly directly into a staking sink, and unlike an emissions-funded sink, it is paid for by parties who expect to earn fees on the other side.
- Fee flow scales with deployments, not headcount. Each new builder-run market adds volume to the same fee pipe without the core team shipping anything, which is the structural argument for why platform economics outgrow product economics here.
- Demand is usage-linked rather than narrative-linked. A builder buying HYPE to post a bond is buying access to a revenue stream, which is a materially different buyer profile from speculative spot demand.
The caveat is that all three of these run in reverse. Bonds unwind, and a slot that stops being profitable returns its stake to float. Sink strength is a function of how many builders find deployment profitable at prevailing fee levels, not a fixed floor.
Truth Ventures’ tests for early-stage investments
Truth Ventures wants to invest in underlying architecture and infrastructure for future blockchain activities, so Datta is more interested in whether a team’s roadmap depends on their own effort or other people’s.
“If every new project or use case requires founders to build it, it usually means that I’m underwriting a company product with various tokens attached. Unfortunately, in these cases, third parties can often extend the thing without permission, so the team’s output loses value. This is why the biggest winners (technology winners) in this particular space aren’t necessarily the applications that everybody can name. Instead, it’s the layer that goes underneath the applications that supports their functionality.”
However, Datta is clear that his approach to investing in crypto infrastructure is not risk-free. The code remains closed-source, so what runs beneath the interface cannot be independently verified, and settlement depends on authorised oracles posting accurate results. For a fully collateralised binary that resolves to 0 or 1, the oracle is the entire trust assumption. There is no price path to arbitrate, only a final call, which makes the reporting layer the single highest-value attack surface on the system.
Competition compresses margins from the other side. Deployment rights are a replicable design; nothing stops a rival venue from offering the same permissionless primitive at a lower bond, so the defensibility sits in liquidity depth and builder switching costs rather than the mechanism itself.
“Ultimately, the platforms that are going to win are those that build reputations among builders,” explains Datta. “If builders repeatedly want to use the same token-based platforms, that should lead to long-term infrastructure network effects that existing players like Hyperliquid can leverage.”
About Varun Datta
Varun Datta is a venture capitalist and CEO of Truth Ventures, a global Web3 venture capital firm backing early-stage companies across Web3 and decentralised technologies. With more than 15 years of experience in venture capital and emerging technology, his work centres on identifying founders and technologies with the potential to shape how decentralised markets develop.
His investment interests span areas such as decentralised AI, DePIN, privacy and blockchain infrastructure. Through Truth Ventures, Varun combines early-stage investment with strategic guidance, working alongside portfolio companies as they develop their technology, expand into new markets and build for long-term growth.