Understand this story
On November 13, 2026, Jakub Kratochvil argued that, in the age of agents, testing software alone is no longer enough; quality assurance must also test the agents themselves. A report published by heise online the same day explained that Agentic QA can use AI agents to quickly reproduce defects, analyze components, generate regression tests, and propose solutions. The corresponding testing scope must cover agents’ planning, tool use, and dynamic decision-making, accompanied by organizational adjustments and clear guardrails. The report also noted that slow organizational approval processes are the main obstacle. There are currently no other qualifying developments.
AI-generated from public reports
Latest development · Report
Agentic QA: Why the Biggest Obstacle Isn’t the CodeDevelopments
1 confirmed developments
Jakub Kratochvil Argues That Software Testing Should Also Test Agents
Source:heise online
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 13
heise online
Agentic QA: Why the Biggest Obstacle Isn’t the Code
Agentic QA enables AI agents to quickly reproduce defects, analyze components, generate regression tests, and propose solutions, but slow organizational approval processes have become the main obstacle. Quality assurance must expand from testing software to testing agents’ planning, tool use, and dynamic decision-making, accompanied by organizational adjustments and clear guardrails.
AI Trends