Answers are grounded in this paper and cite their sources.
Smart cities are moving to depend on big data and AI-focused data streams for key decisions, yet existing frameworks are often based on centralised architectures that are manipulable, with poor auditability, and deepfakes present a huge risk. ChainProv-City is an authority-based blockchain provenance system that provides trust and accountability in a smart city surveillance context. Unlike traditional approaches based on anonymous consensus, the model introduces institutional authorities into the validation process, ensuring that validation aligns with real-world governance frameworks. Each event is anchored in a device-signature commitment, adjudicated by k-of-m identity-bound agency votes with timeouts and a safety escalation route, and recorded in a hash-linked provenance trail; membership changes are managed by a separate governance quorum. Security is analysed in an explicit adversary model, and a representative deepfake case is simulated as executable tests of governance under forged attribution, not deepfake detection. The Solidity prototype passes 40 automated tests. Registration consumes about 265k gas, a vote about 82k gas, throughput is limited by the block gas limit to about 56 events/s (66 with batching) and the median decision latency is below 4.0 s (excluding human review time); single-node Hardhat figures, i.e., best-case bounds.
No takes yet. Share an insight, caveat, or question.
Farhan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: