Kommentar: Die KI-Transformation scheitert am Engpass Mensch
Unternehmen fragen sich, warum der KI-Effekt ausbleibt – und investieren weiter in leistungsfähige Modelle. Dabei liegt es nicht an der Technik, sondern am Menschen. Warum ändern wir das nicht?
The question on every company's mind is whether their artificial intelligence (AI) investments are paying off. According to the "State of AI Report 2026" by McKinsey, 49 percent of German respondents have already scaled AI across their organizations. They have moved past the pilot phase and automated first processes using AI agents.
However, 43 percent cannot quantify their contribution to the result. So, why does the AI effect remain elusive? A recent report by Anthropic sheds some light. The company's Claude developers measured how their own AI changed internal workflows and the extent of AI influence. They analyzed personnel data, calendars, Slack communication, hiring logs, and 17 interviews.
While the data is not independently verifiable, it reveals a problem: not the technology is the bottleneck, but the human. Users are the choke point for the agents. Despite AI agents being six times more productive than humans, they spend around 80 percent of their interactive sessions waiting for human inputs. "The users are the bottleneck for the agents," the report concludes.
Written by urgent.news from Handelsblatt's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
