Bottom line: Enterprise-grade AI agents that orchestrate workflows across multiple systems are required to translate AI ambitions into operational value and meet regulatory requirements.
36 percent of German companies now use AI – nearly twice as many as a year ago. Yet 95 percent have so far achieved no measurable business impact, because pilot projects are not being converted into reliable operational systems.
The adoption pressure is real: From 2023 to 2024, AI usage in Germany has doubled from 20 to 36 percent. With this proliferation, AI is shifting from an experimentation phase to an operational necessity – and thus also into a cost centre. Delays in scaling lead to measurable service delivery failures and rising costs.
Despite billions in investment in generative AI, 95 percent of companies have so far failed to realize tangible business results. The reason does not lie in a lack of motivation or partners – Microsoft, Cisco, Adobe and Google actively support efforts. The central problem: companies cannot convert AI pilot projects into orchestrated systems that work reliably in complex, regulated industries. Chatbots and copilots improve individual points of efficiency, but cannot handle multi-stage, cross-system workflows with real-time decision-making and validation. Here either the automated process fails or human intervention becomes necessary.
In practice, AI in German firms focuses on customer service (88%) and marketing/communications (57%), mostly through chatbots for call volume reduction and FAQ answering. These measures brought incremental improvements but rarely transformed the customer experience – also because acceptance remains limited: 53 percent of consumers see benefits in GenAI chatbots, while 28 percent continue to prefer contact with humans.
An additional obstacle arises in regulated sectors such as banking, insurance, healthcare and the public sector. Current GenAI models are not built for role-based access, audit trails or guarantees of approved policy conformance. Business-critical processes – financial transactions, policy changes, administrative services – cannot therefore be entrusted to them. These sectors rely on deeply integrated systems of record, whose replacement is unrealistic. New technologies that circumvent governance frameworks create compliance risks.
The solution lies in enterprise-grade AI agents that orchestrate workflows across CRM, contact centres and core platforms – not working in isolation. They guide users through multi-stage processes, retrieve information, validate it and trigger actions across integrated systems. Only with this architectural reassessment can AI be transformed from experiment to operational reality with measurable return.
Source: www.it-daily.net · Published 10 June 2026
Lumi AI News — AI-assisted curation pursuant to Article 50 EU AI Act. Paraphrasing and classification by Lumi News Pipeline v1.6.5.