Cloud‑Powered AI & Blockchain: Key Takeaways from UCloud TIC 2018

Cloud Enables AI Across Industries

The conference opened with a discussion on how cloud computing serves as the foundational infrastructure that empowers AI to tackle real‑world problems in customer service, education, and autonomous driving.

UCloud AI expert Song Xiang described the technical challenges of AI deployment and presented the UAI PaaS platform, which offers environment packaging, automatic scheduling, load balancing, and massive resource sharing to accelerate AI research and production.

Pan Zheng from Yushe Technology explained real‑time deep‑learning on embedded devices, emphasizing the trade‑offs between latency, cost, and accuracy for vision‑based autonomous driving, and highlighted the suitability of FPGA‑based solutions.

UCloud senior R&D director Ye Lideng demonstrated a machine‑learning platform architecture, covering resource management, containerized algorithms, cross‑zone disaster recovery, gray‑release, and parallel scaling.

Fourth Paradigm’s Xing Shaomin addressed challenges in building intelligent customer‑service bots, such as data scarcity and multi‑turn dialogue, and suggested a human‑in‑the‑loop approach to improve accuracy over time.

Chu Shi Technology’s Li Shuguang shared AI applications in education, including speech assessment, homework grading, and handwriting recognition, detailing how NLP, computer vision, and machine‑learning models evaluate grammar, structure, and coherence.

Blockchain Applications and Security

Blockchain was presented as a distributed ledger technology with potential in finance, credit, IoT, trade settlement, and asset management, but its decentralized consensus raises privacy and data‑processing concerns.

ChainNova CEO Dong Ning identified three major pain points for blockchain adoption and introduced the ChainNova BaaS solution, which integrates multiple chains, big‑data storage, and pre‑packaged micro‑services to provide scalable, “out‑of‑the‑box” blockchain platforms.

UCloud incorporated Intel SGX trusted execution environments to protect confidential data in smart contracts, enabling confidential computing on blockchain nodes.

Examples of blockchain use cases included insurance (presented by PICC Group’s Li He), digital‑currency security (UCloud security lead Zong Ze discussed offline wallets and exchange protections), and IoT security (Blockshine’s Meng Xi discussed tamper‑proof, peer‑to‑peer, and anonymity features).

Conclusion

The event concluded that AI and blockchain are still in early development stages, and cloud‑enabled solutions are essential for their practical adoption, requiring a complete industry ecosystem and sustained technological benefits to justify investment.