According to @VitalikButerin, Fusaka prioritizes safety and centers on PeerDAS to enable a live blockchain where no single node must download full block data, a design aimed at robust data availability for scaling. source: Vitalik Buterin on X, Sep 24, 2025. He explained that PeerDAS has each node request a small number of chunks to probabilistically verify that more than 50 percent of chunks are available, after which nodes can download those chunks and use erasure coding to recover the rest. source: Vitalik Buterin on X, Sep 24, 2025. In the first version, full block data must still exist in one place for initial broadcasting and for reconstruction when a publisher releases between 50 percent and less than 100 percent of a block, but these roles are untrusted and require only one honest actor to succeed even if many are dishonest. source: Vitalik Buterin on X, Sep 24, 2025. He added that future cell-level messaging and distributed block building will distribute even these remaining functions. source: Vitalik Buterin on X, Sep 24, 2025. Core developers will be highly cautious on testing, and the blob count will increase conservatively at first before becoming more aggressive over time. source: Vitalik Buterin on X, Sep 24, 2025. He stated PeerDAS is key to L2 scaling and eventually L1 scaling once the L1 gas limit is high enough to place L1 execution data into blobs, framing Fusaka and PeerDAS as central to Ethereum’s scaling roadmap. source: Vitalik Buterin on X, Sep 24, 2025.