Tag
A blogger recounts spotting AI-generated customer service replies from Ocado, using tells like sycophantic agreement and em-dashes, and how complaining loudly led to a human apology. The piece advocates for the AI Resist List and pushing back against poor AI deployments.
A builder shares their system for deploying customer-facing WhatsApp agents in a single prompt, with code-enforced scoped 'actions', rate limits, and message debouncing to safely connect agents to credentialed systems like Airtable, Google Sheets, or MCP servers.
Australia's Commonwealth Bank reversed its decision to replace 45 call centre workers with an AI voice bot after call volumes rose, illustrating Gartner's finding that only 14% of customer service issues are fully resolved through self-service.
New York City's DCWP proposed rules that would require AI chatbots to disclose themselves, provide a one-step path to human agents, and ban customer service 'doom loops', with fines up to $3,500 per violation — potentially making it the first US city to enforce such protections.
ALTK-Evolve introduces a learned-memory approach that tops the τ²-bench leaderboard on customer service tasks, achieving state-of-the-art results without relying on a frontier model.
Bold Agent Kit 分析约 14,000 通来电后发现，尽管 AI 接线员接听更多电话，人类接待员的转化率仍高出约 4 倍，且 AI 处理的电话中无效挂断率高出 700%，表明 AI 自动化在替代人工的同时尚未在效果上胜出。
Bland 推出可从网站链接自动生成的 iMessage 业务 AI 代理，支持品牌认证、多模态图片识别、支付与预约等前台工作流，Apple 的 Business Register 成为验证发送方的入口。
Bland 推出 iMessage 智能体功能，可在30秒内将任意商业网站转化为带品牌标识的短信客服线路，支持实时抓取价格、库存并接受支付。
多位语音AI公司高管在HumanX大会上表示，尽管全双工模型和语音AI投资热潮不断升温，语音AI仍未迎来自己的“ChatGPT时刻”，低延迟推理、ASR准确率和对话透明度仍是关键挑战。
Meta has replaced human customer support on Instagram with AI-only support for paying customers, but the AI support reportedly doesn't work at all, leaving users unable to report issues or obtain receipts.
Patio11 分享了一则轶事：用户被困在 Verizon 的账单纠纷中，转而使用 Codex 桌面版 AI 来处理与顽固官僚机构的沟通，展示了 LLM 在争议解决中的实用价值。
A practitioner describes how their AI voice agent hallucinated a delivery date and then doubled down on it across multiple turns as its own output accumulated in context, and how separating tool-returned facts from the agent's own utterances fixed the issue more reliably than prompting the model to treat its answers as provisional.
Customer Service AI for Etsy is a Product Hunt launch that generates polished, shop-voiced replies to Etsy buyer messages in seconds, with a pending-reply to-do list and shop memory, human-in-the-loop before sending.
A Verge report examines how AI hallucinations are worsening customer behavior, with restaurant, retail, and park workers reporting customers who trust ChatGPT over experts on allergens, wine, and travel itineraries.
An opinion piece exploring the accountability gap when AI systems make or block service decisions — such as application denials or flagged transactions — and arguing that important AI-assisted decisions should come with a clear human review and appeal process.
The article explores Cresta's AI agent for conversation intelligence, focusing on its potential to automate customer interactions while providing integrated analytics for quality assurance and coaching, and questions whether this unified platform simplifies operations compared to separate tools.
A tweet warns that businesses are unprepared for the rise of agentic traffic, with less than 1% of consumers currently using agents, but this is expected to increase significantly, bringing new challenges in managing customer interactions.
@wonder is building an AI agent to automate meal planning and delivery, integrating LangSmith for debugging and scoring using LLM-as-judge.
An AI agent named Muse sent an auto-reply confirming availability during a pickup, leading to a negative rating when the user wasn't present, prompting a discussion on improving auto-replies.
The article discusses the risks of AI voice agents making promises to customers, such as scheduling callbacks or refunds, which could create unexpected work for other teams and highlight ethical concerns in automation.