https://manuel.bernhardt.io/tags/ai/Recent content in Ai on Manuel BernhardtHugoen-usSun, 01 Nov 2026 19:43:05 +0100https://manuel.bernhardt.io/posts/2026-11-01-slopstructions/Sun, 01 Nov 2026 16:26:38 +0200https://manuel.bernhardt.io/posts/2026-11-01-slopstructions/<p>These days I&rsquo;m quite busy handling cloud infrastructure and security for a client. This role is quite demanding with everyone using AI to do their job: on one hand, software ships faster; on the other hand, CVEs get discovered faster and need to be remediated much faster as a result.</p> <p>One thing that is starting to seriously annoy me is the rather new phenomenon of people with a complete lack of competency in infrastructure &amp; security handing me over detailed step-by-step instructions of what to do in order to get something from me. I&rsquo;m calling these <em>slopstructions</em>: seemingly correct instructions (or so it may apper to the <strike>author</strike> prompter) based on a fundamental misunderstanding and lack of knowledge of the process, technology, security implications, etc.</p>