Zattoo streaming prototypes cut live latency to two seconds via QUIC

Reducing live latency to two seconds via Media over QUIC is a critical technical milestone for interactive streaming and real-time AI applications like live dubbing. By moving toward sub-three-second delays, Zattoo is positioning its infrastructure to support high-stakes live sports and synchronized social viewing that current HLS or DASH implementations struggle to maintain. The integration of dedicated NPUs in future hardware suggests a strategic shift toward edge computing, reducing cloud processing costs for complex tasks like natural-language discovery. Watch for whether Zattoo moves these MoQ implementations into its commercial Zattoo Modular and End2End portfolios following operator feedback at IBC.

Zattoo's Media over QUIC work arrives as the IETF MoQ working group pushes toward standardization and competing low-latency delivery approaches gain commercial traction. The IETF's MoQ Transport draft has attracted contributions from major streaming infrastructure players, and Ericsson and Nokia are diverging on AI-RAN architectures that could underpin future low-latency video delivery over mobile networks, with Nokia building its entire Layer 1 RAN on Nvidia's CUDA platform and GPUs while Ericsson reserves GPU acceleration for forward error correction alone. That architectural split matters for streaming providers like Zattoo because the radio access network's processing model directly affects end-to-end latency budgets for live video. Meanwhile, a cluster of announcements in June 2026 signaled a real shift from AI research to commercial AI-driven network automation across telcos, with Ericsson launching its AI in RAN commercial software subscription claiming up to 20% higher downlink throughput and Verizon disclosing that its 60,000-site vRAN now applies agentic AI to configuration changes and network optimization.

On the business side, Zattoo's prototype strategy reflects broader industry movement toward AI-embedded operations platforms rather than standalone tools. Nokia combined with AWS and Databricks to build a telco AI control layer around its Autonomous Network Fabric, positioning the fabric as an orchestration layer that consumes data, applies models, and triggers actions across radio, core, transport, and service domains. Nokia claims operators using its autonomous networks portfolio are achieving automation rates higher than 90 percent and service delivery times of four hours or less. The parallel for streaming platforms is clear: Kaltura AI Cloud TV features and real-time subtitle translation prototypes represent the same pattern of embedding AI agents directly into the delivery and operations stack rather than bolting them on as separate services.

Technical benchmarks from adjacent deployments underscore the latency and efficiency gains Zattoo is targeting. Nokia and Google Cloud unveiled six Gemini-powered AI agents at DTW Ignite 2026 capable of slashing network problem-solving times by 50% to 80%, with agents covering event triage, anomaly reasoning, and remediation recommendation. Nokia plans to launch its agentic platform in Google Cloud Marketplace in September 2026, with operators deploying the router and event triage agents via Nokia Assurance Center. For Zattoo, the relevance is twofold: agentic AI frameworks are maturing fast enough to support real-time streaming operations at scale, and the sub-two-second latency targets Zattoo demonstrated with MoQ align with the network-side automation speeds these platforms now deliver. The convergence of low-latency transport protocols and embedded AI agents suggests that streaming providers adopting both simultaneously will hold a structural advantage in live sports and interactive content delivery.