By Amir Hassan, Blockchain Technology Correspondent
GIBRALTAR — In what industry veterans are calling the most significant architectural shift since the launch of the beacon chain, Polkadot has officially activated the Join-Accumulate Machine (JAM) on its mainnet. As of 10:00 UTC today, May 8, 2026, the legacy Relay Chain has begun its final sunset phase, handing over the reins of the network’s security and coordination to a new, decentralized “multi-core” engine designed to provide raw, generalized computation at a scale previously reserved for centralized cloud providers.
The activation follows two years of rigorous development and the successful distribution of the 10 million DOT “JAM Implementer’s Prize,” which saw seven distinct teams—including developers from Parity Technologies, Chainsafe, and several independent research collectives—deliver compatible client implementations. This multi-client approach has effectively decentralized the protocol’s core logic, a feat that Gavin Wood, Polkadot’s founder, described during today’s launch event as “the realization of a transactionless, pure compute layer.”
The Death of the Relay Chain, the Birth of “Services”
For years, Polkadot operated as a “chain of chains,” where parachains (layer-1 blockchains) auctioned for slots to connect to a central Relay Chain for shared security. While revolutionary in 2020, this model was often criticized for its rigidity and the technical overhead required to maintain a full blockchain just to run a simple decentralized application.
JAM fundamentally alters this relationship. By replacing the WebAssembly (Wasm)-based Relay Chain with the Polkadot Virtual Machine (PVM)—a high-performance architecture built on the RISC-V instruction set—JAM allows for “Services” to run directly on the network’s cores without the need for a dedicated blockchain structure. These services can range from traditional smart contract environments and ZK-rollup sequencers to decentralized AI training modules and distributed storage indexes.
“We are no longer just securing other blockchains,” said Robert Habermeier, co-founder of Polkadot, in a statement to BitcoinsNews. “We are providing a substrate for generalized computation. Today, the first three ‘Exo-Services’—highly specialized compute modules that don’t follow the traditional blockchain heartbeat—have successfully synchronized with JAM’s cores. This isn’t just Polkadot 2.0; it’s the end of the ‘chain-centric’ era.”
Technical Milestones: 1 Trillion Gas and SAFROLE
The technical specifications of the JAM upgrade are staggering. By utilizing Agile Coretime—a system that allows developers to purchase compute power on-demand or in bulk—the network is now capable of processing an estimated 1 trillion EVM-equivalent gas per second across its distributed cores. This throughput puts Polkadot in direct competition with high-performance monolithic chains like Monad and Solana, while maintaining the rigorous security of a sharded architecture.
Central to this performance is the SAFROLE consensus protocol, a SNARK-based block production algorithm that ensures near-instant finality and robust resistance against MEV (Maximal Extractable Value) exploits. Unlike previous iterations, SAFROLE allows JAM to maintain a “transactionless” core, where state transitions are proven via zero-knowledge proofs before being “accumulated” into the global state.
Market Reaction and the “DOT” Utility Shift
The market has reacted with cautious optimism. DOT, the native utility token of the ecosystem, has seen a 14% increase in trading volume over the last 24 hours, currently hovering around the $42.50 mark—a significant recovery from the mid-2024 lows. The shift in utility is the primary driver; under JAM, DOT is no longer just for staking and governance, but serves as the direct “fuel” for Coretime. Large-scale enterprises and AI labs are reportedly accumulating DOT to secure long-term compute bandwidth, a move that parallels the way Web2 companies reserve server capacity in AWS data centers.
“The demand for verifiable compute is the new oil of the 2026 economy,” noted Sarah Chen, Lead Analyst at BlockTrend Research. “With JAM, Polkadot has successfully pivoted from being a niche interoperability play to a foundational infrastructure provider for the decentralized web. The fact that they achieved this while maintaining a multi-client ecosystem is a massive win for decentralization.”
Analysis: Can Polkadot Overtake the Parallel EVM Giants?
The activation of JAM comes at a time when the “Parallel EVM wars” are at their peak. Monad, which dominated the narrative throughout late 2025, has focused on optimizing the existing Ethereum developer experience. Polkadot’s approach with JAM is more radical—and potentially more rewarding. By moving away from the blockchain-as-a-container model, JAM allows for heterogeneous services that can interoperate at the speed of the underlying hardware cores.
However, challenges remain. The complexity of writing “Services” for the PVM remains higher than deploying a standard Solidity contract. To combat this, the Web3 Foundation today announced a $50 million “Developer Abstraction Fund,” aimed at creating tooling that allows Ethereum and Solana developers to port their logic to JAM services with minimal friction.
The Road Ahead: Elastic Scaling and Beyond
Looking forward, the Polkadot roadmap for 2026 includes the rollout of “Elastic Scaling,” which will allow a single Service to dynamically spread its workload across multiple JAM cores during periods of high demand. This feature, expected to go live in Q4, would effectively eliminate “gas wars” by allowing the network to expand its capacity in real-time.
As the first blocks of the JAM era are finalized, the industry is watching closely. If Polkadot can prove that a transactionless compute layer is more efficient than the traditional “chain-of-chains” model, it may very well set the standard for the next decade of decentralized infrastructure. For now, the successful launch of JAM marks a bold gamble on a future where the word “blockchain” might just become an implementation detail of the larger World Computer.
1 trillion EVM-equivalent gas per second is an insane claim. need to see actual mainnet benchmarks before getting excited, but the architecture is compelling
SAFROLE using SNARK-based block production with near-instant finality is the kind of thing people said was impossible on sharded arch. gavin actually shipped it
lol remember when people called dot dead in 2024. seven independent client implementations and RISC-V based PVM later…
DOT at $42.50 after mid-2024 lows feels like the accumulation phase people will look back on. enterprises buying coretime is real demand, not speculation
goodbye parachain auctions, you were fun but honestly leasing a chain just to deploy a dapp was always clunky. services model makes way more sense
the $50M dev abstraction fund is smart. biggest risk for JAM adoption is tooling, not tech. if sol devs can port easily this gets real traction fast
elastic scaling in Q4 eliminating gas wars would be massive. monad better watch their back