AI customer service is not ready for prime time

· Source: Semafor · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Corporate Strategy & Leadership, Government & Public Sector · Depth: Fundamental Awareness, extended

Summary

The AI landscape is marked by significant challenges and rapid developments, as highlighted by recent reports. AI customer service, despite advanced voice models like ElevenLabs', struggles with user experience, recognition, and lag, leading to low trust and preference for human agents. Companies are grappling with AI's return on investment (ROI), with IBM's Gary Cohn noting "massive over-investment" and JPMorgan observing token costs exceeding some employee salaries. This has led to reduced AI spending and a focus on measurable automation. Meanwhile, Microsoft launched Scout, an OpenClaw-powered AI assistant for enterprise, while OpenAI introduced new corporate tools, acknowledging Anthropic's enterprise success. Concerns over AI safety are rising, with Anthropic calling for a development slowdown due to "recursive self-improvement" risks, and AI CEOs jointly warning about bioweapon threats. Public opposition to AI infrastructure is growing, exemplified by Monterey Park's ban on data center construction, driven by energy demands and privacy concerns, which also fuels a tech industry bet on PC-based AI.

Key takeaway

For executives evaluating AI investments, prioritize solutions with clearly measurable ROI, especially in areas like sales or customer experience, to justify token spending. Be aware that public opposition to AI infrastructure is rising, and user trust in AI customer service remains low, requiring careful deployment strategies. Engage with policy discussions around AI safety and regulation, as calls for development slowdowns and wealth-sharing models are gaining traction, impacting future operational costs and public perception.

Key insights

AI's rapid advancement brings high costs, ROI measurement difficulties, and significant safety/societal challenges, necessitating careful deployment and regulation.

Principles

Method

Companies can mitigate AI token costs by focusing on automating measurable tasks and optimizing for performance plateaus, while also considering local processing for sensitive data.

In practice

Topics

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Tech Journalist, Executive, Policy Maker

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Semafor.