Harmonic cOS SensAI launch targets $1 billion tier 2 broadband market

This move signals a shift in the broadband vendor landscape as Harmonic pivots toward a pure-play software model following the sale of its video business to MediaKind. By integrating Google's Gemini, Harmonic provides smaller ISPs with the sophisticated network automation tools previously reserved for tier 1 giants like Comcast. This democratization of AI infrastructure allows mid-market providers to defend margins against rising competitive pressure and multi-year price locks from larger rivals. The industry should monitor whether the National Content & Technology Cooperative integrates SensAI into its own AI Center of Excellence for its 700-plus member operators.

Harmonic's entry into AI-driven network operations places it in direct competition with established telecom automation vendors that have been aggressively expanding their agentic AI portfolios. Nokia, for instance, launched six Gemini-powered AI agents for telco network troubleshooting in partnership with Google Cloud at DTW IGNITE 2026, targeting alarm triage, KPI analysis, anomaly detection, and remediation for large operators. Nokia claims these agents can reduce network problem-solving times by 50% to 80%, with plans to make the platform available on Google Cloud Marketplace in September 2026. The parallel use of Google's Gemini model by both Harmonic and Nokia underscores how the hyperscaler's AI stack is becoming a common foundation across network automation, though Harmonic targets a fundamentally different customer base of tier 2 and 3 broadband providers rather than tier 1 mobile operators. The broader AI-RAN and network automation market is consolidating around two competing architectural philosophies that will shape vendor selection for years. Ericsson launched its AI in RAN commercial software subscription on June 11, 2026, claiming up to 20% higher downlink throughput across more than 15 live deployments, while Nokia and Nvidia announced a GPU-accelerated AI-RAN partnership with Indosat Ooredoo Hutchison in Indonesia on June 8. Verizon disclosed that its 60,000-site vRAN network is now applying agentic AI to configuration changes and service assurance, publicly calling for industry-wide interoperability standards. The TM Forum's Autonomous Networks L4/5 roadmap and 3GPP 6G standardization process are expected to incorporate agentic AI interoperability as a core requirement, which could eventually benefit Harmonic's multi-vendor positioning among smaller operators who lack the resources to manage proprietary automation stacks. On the technical front, the divergence between GPU-accelerated and custom-silicon approaches to network AI has direct implications for Harmonic's cloud-native SaaS model. Light Reading reported that Ericsson and Nokia are diverging sharply on AI-RAN architecture, with Nokia running all Layer 1 functions on Nvidia GPUs while Ericsson limits GPU use to forward error correction. Nokia's anyRAN platform, built on Nvidia's Aerial computing environment, has reportedly delivered spectral efficiency improvements exceeding 20% with plans to reach 50% by 2027. Harmonic's approach sidesteps this hardware debate entirely by delivering AI capabilities as a cloud service, which aligns with the economic constraints of its target market. Smaller broadband operators typically lack the capital expenditure budgets for GPU infrastructure or dedicated AI engineering teams, making a subscription-based model powered by Gemini a pragmatic path to automation parity with larger competitors. As the industry evolves, Samsung agentic AI networks are also beginning to target autonomous Level 5 telecom operations, further highlighting the rapid pace of innovation in this sector.