AI & Automation
AI Virtual Agents Reshape Contact Center Resolution Standards: From Transfer to End-to-End Resolution
In 2026, the performance metrics for AI virtual agents in contact centers are undergoing a fundamental shift: from traditional call transfer rates to end-to-end resolution rates. The industry no longer measures success by reducing the number of human interactions, but rather by whether customer issues are truly resolved. Modern AI virtual agents, powered by large language models, retrieval-augmented generation, and real-time system integration, possess cross-channel contextual memory, intelligent handoff, and multi-step operational execution capabilities, enabling them to autonomously resolve issues within enterprise-defined guardrails. Based on Metrigy research data, this article analyzes the three core capabilities of a resolution-first strategy and explores their strategic impact on customer satisfaction, employee burnout, and operational costs.
Agentic Workforce Management: A Paradigm Shift from Manual Monitoring to Autonomous Operations
Workforce Management (WFM) has relied on manual intervention at every step for three decades, whereas Agentic WFM achieves continuous monitoring, proactive alerting, and automated execution of routine tasks by connecting data, AI, and automation. Based on McKinsey's 2025 AI report (62% of enterprises remain in early experimental stages) and BCG analysis (more jobs are being reshaped rather than replaced), and incorporating real-world cases such as DraftKings, the article elaborates on the three-stage evolution of Agentic WFM, its practical forms, and its profound impact on the role of WFM professionals—shifting from operational execution to orchestration, judgment, and accountability.
Uber cuts 10% of customer service team positions, accelerating AI transformation
Uber cuts 10% of customer service positions and mandates remote employees to return to offices to streamline operations and embrace AI. The company describes this as a business restructuring to pave the way for future AI success. Analysts note that many enterprises use AI as a pretext for layoffs, actually adjusting cost structures. Forrester predicts AI will eliminate half of customer service roles by 2030, but cases like Klarna show that some companies may rehire due to poor AI implementation.
Bank of America upgrades customer service employee AI tool EricaAssist, adding generative AI real-time suggestion feature
Bank of America announced on Tuesday that it has introduced generative artificial intelligence capabilities to EricaAssist, the AI assistant used by its customer service employees, allowing it to provide personalized guidance within seconds to help employees respond to customer needs more quickly and efficiently. The bank said the new feature is now live and plans to expand it to more service scenarios and business lines within the year.
Clear human agent transfer paths can enhance consumer trust in AI customer service
A Five9 survey of 3,000 consumers found that clear human agent transfer paths significantly boost consumer trust in and willingness to use AI customer service. Even when the AI technology's effectiveness remains unchanged, trust doubles when a human option is present. Additionally, 41% of consumers said they would reduce their usage if a company used AI customer service, and this rose to 53% when no human option was available. Experts emphasize that disclosure and human support are key to building trust.
Third-party generative AI tools outperform brand-owned chatbots in customer service
A Gartner survey of more than 3,500 B2B and B2C customers found that customers are three times more likely to use third-party generative AI tools for customer service than brand-owned chatbots. Over the past year, usage of third-party generative AI tools has doubled, while usage of enterprise-provided chatbots has shown no statistically significant growth since 2022. Analysts note that brands need to rethink their AI customer service strategies and recommend transforming the entire digital experience into an AI-driven conversational interface.
AI Drives Shrinkage in Customer Service Labor Market: Job Structures Accelerate Differentiation
According to the latest Forrester research, as AI expands its role in contact centers, the U.S. customer service labor market is experiencing contraction, with job postings down about 10% from pre-pandemic levels. Over the next five years, lower-level automatable positions will decrease, while new knowledge-based roles created by AI will increase.
How Customer Experience Leaders Manage the Workforce in the AI Era
As AI becomes increasingly prevalent in contact center applications, customer experience leaders are undergoing profound changes in workforce management. Based on a panel discussion at Customer Contact Week in Las Vegas, this article synthesizes practical perspectives from executives at Talkdesk, InfoPay, Five9, and other companies, emphasizing the importance of transparent communication, employee engagement, and rational technology selection.
The Gap Behind the Success of AI Projects: Customer Experience Leaders Face Reality
A survey released by Laivly shows that although most CX leaders claim AI projects are successful, 53% exceeded budget, 43% were delayed or stalled, and 28% of leaders attributed revenue loss to AI's inability to handle customer complexity. Gartner analyst Ian Elliot notes that the disconnect between leadership perception and reality stems from market pressure, misaligned personal incentives, flawed performance metrics, and insufficient accuracy of AI tools.
The initiative in customer relationships always belongs to you
Decagon, through tools such as Agent Operating Procedures and Duet, enables companies to deploy AI agents for complex customer service workflows within three weeks, whereas previously using competitors required nine months. Multiple enterprises (such as a global airline, a music streaming platform, a social network, and a health insurance company) have achieved rapid deployment and high success rates. The article emphasizes that enterprise software should activate rather than appropriate customer relationships, and Decagon's glass-box design gives operators control over iteration speed instead of relying on vendors.