Generative AI Cuts Customer Service Costs by Up to 68%, But the Fully Autonomous Vision Isn't Materializing, Epignosis Insights Reports

Epignosis Insights Research Desk today released "Generative AI in Customer Service: Promise Versus Real-World Performance," a report assessing how generative AI is actually performing in customer service, as distinct from how it is being marketed. The gap is substantial. Gartner has told the market that agentic AI will autonomously resolve 80% of common customer service issues by 2029 while cutting operational costs by 30%, and in the same research cycle has warned that more than 40% of agentic AI projects will be canceled by the end of 2027 due to unclear business value and inadequate governance.

Get Full Access of Report @https://epignosisinsights.com/report/generative-ai-customer-service-promise-vs-performance

"Generative AI delivers real, measurable value in customer service," the report concludes, "but almost never in the unqualified, fully autonomous form vendors describe in press releases."

Klarna's Rise, Reversal, and Rebalancing

Klarna remains the industry's defining case study, the report finds, because the company was transparent about both its initial success and its subsequent correction. Launched in February 2024, Klarna's AI assistant processed 2.3 million conversations in its first month and cut average resolution time from 11 minutes to under 2 minutes, publicly framed as equivalent to replacing 700 full-time agents. By May 2025, CEO Sebastian Siemiatkowski acknowledged the company had focused too much on efficiency and cost and that the result was lower quality, and Klarna began rehiring human agents for premium accounts and complex, emotionally sensitive cases. By the third quarter of 2025, Klarna's own earnings disclosures described the assistant as automating work equivalent to 853 full-time agents and generating $60 million in cumulative savings, a larger automated footprint than the original claim, achieved after the company learned which categories of interaction AI could and could not safely own.

"Agent Washing" and the Governance Problem

Gartner's own research provides the clearest documented evidence of a structural gap between marketing and deployed capability, the report notes. Of the thousands of vendors currently marketing agentic AI customer service products, Gartner estimates only approximately 130 offer genuine agentic capability, systems that can reason across multi-step problems, coordinate across channels, and take autonomous action within enterprise systems. Gartner labels the remainder "agent washing," and industry analysis following up on that research estimates these relabeled systems typically handle only 20 to 30% of interactions in production, far short of the 80% autonomous-resolution figure Gartner itself projects for 2029.

Consumers Want Transparency and an Escalation Path

Independent survey research from Salesforce, based on responses from 16,585 consumers and business buyers worldwide, quantifies a consumer trust problem underneath the deployment statistics, according to the report. Seventy-two percent of consumers say they trust companies less than they did a year ago, 75% say it is important to know whether they are communicating with an AI agent rather than a human, and 45% say they are more likely to use an AI agent at all if a clear escalation path to a human exists. The report notes that Klarna's course correction toward a hybrid model aligns closely with what this survey data says customers actually want.

Regulators Move From Watching to Enforcing

U.S. regulators have moved from watching this space to actively enforcing against it, the report finds. The FTC's Operation AI Comply, launched in September 2024, has produced at least eight settled enforcement cases targeting companies that overstated AI capability, including a $193,000 settlement with an "AI lawyer" chatbot service finalized in February 2025. In September 2025, the FTC broadened its focus with a formal inquiry into seven major AI chatbot providers, and a December 2025 executive order directed the FTC to clarify that misrepresenting the likelihood of AI hallucinations to consumers can constitute deception.

"The pattern that recurs across McKinsey's research, Klarna's disclosed metrics, and Salesforce's consumer survey data is consistent: generative AI performs reliably on structured, low-emotional-stakes interactions, and underperforms on complex, ambiguous, or emotionally charged interactions where customers specifically want to reach a human," said Abhijith Nair, Senior Research Manager at Epignosis Insights and lead author of the report.

The report cites McKinsey research showing companies implementing AI in customer support have reduced average cost per interaction by 68%, from $4.60 to $1.45, concentrated in high-volume, low-complexity interactions such as order status, password resets, and appointment scheduling. Epignosis Insights' outlook argues that the more useful planning question for organizations is not how much of customer service AI can replace, but which specific interaction types AI can safely own, and what the escalation path to a human needs to look like for the rest.

Epignosis Insights
703 Kumar Corporate Building, Pune-411028, India
sales@epignosisinsights.com
Website: www.epignosisinsights.com

About Epignosis Insights
Epignosis Insights is a market research and publishing consultancy based in Pune, India, providing brand health tracking, customer experience research, and market intelligence to organizations across financial services, FMCG, automotive, and industrial sectors. The firm's Research Desk compiles and analyzes data from government agencies, industry associations, corporate disclosures, consulting firms, and credible news sources to produce independent, data-anchored market reporting.

This release was published on openPR.