{"id":1742,"date":"2026-07-18T20:58:56","date_gmt":"2026-07-18T19:58:56","guid":{"rendered":"https:\/\/thomas-sokolowski.com\/en\/?p=1742"},"modified":"2026-07-19T07:41:18","modified_gmt":"2026-07-19T06:41:18","slug":"ai-the-mediocrity-amplifier","status":"publish","type":"post","link":"https:\/\/thomas-sokolowski.com\/en\/ai-the-mediocrity-amplifier\/","title":{"rendered":"AI: The Mediocrity Amplifier"},"content":{"rendered":"\n

What can I say? <\/strong>The results weren’t terrible. But they weren’t particularly original, distinctive, or memorable either. Like so much AI-generated content, the designs stood out mainly for their algorithmic mediocrity.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/div>\n\n\n\n

But how could it be any different? The neural networks behind generative AI tend to treat the average of the average as the safest and most correct solution. More advanced models may improve the results, but they will not solve this basic problem anytime soon.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/div>\n\n\n\n

Deep learning works by identifying patterns that appear repeatedly across large amounts of data. Patterns that occur very often are more likely to be treated as reliable or correct. As a result, generative AI naturally tends to produce solutions that are statistically common.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Please do not get me wrong: Average is not necessarily bad.<\/strong> Nobody can be excellent at everything, so an average result may be perfectly acceptable or even genuinely useful for many people.<\/p>\n\n\n\n

Unless, of course, you hire a specialist. In that case, you might reasonably expect something above average. As a \u00abprofessional\u00bb designer, I have never considered it my job to create merely average solutions. Sure, average might sometimes have been enough. But clients usually hire me to do something they cannot do themselves, not even with the help of AI.<\/p>\n\n\n\n

This is why I find it hard to get excited about what Claude currently produces as \u00abdesign\u00bb. For my work, it is simply not useful.<\/p>\n\n\n\n

Successful designers have always been those who did not simply offer what everyone else offered. They discovered what other people were not particularly good at and then became extremely good at exactly that.<\/p>\n\n\n\n

But back to the original point: For many people, an average solution is apparently good enough. And honestly, why not?<\/p>\n\n\n\n

Most people also live in an average apartment and lead a fairly average life. Mediocrity is everywhere, and AI is giving it another helpful little push. In all these cases, including design, AI is mainly a channel that processes the average opinions and average data of our world into new average opinions and average data. It is, in other words, a mediocrity amplifier.<\/strong><\/p>\n\n\n\n

Design, on the other hand, lives through originality, contrast, personality, and sometimes uniqueness. So how can we train neural networks to consider not only the most common statistical patterns but also independent and innovative design ideas? How can we teach them creative judgment?<\/p>\n\n\n\n

I do not know. Perhaps we will figure it out one day. But I seriously wonder why we should bother.<\/p>\n\n\n\n

Even if AI occasionally produced something truly outstanding, most users probably would not notice. And they certainly would not appreciate it. Most people do not want something extraordinary. They want something familiar, predictable, and average. That is what they expect. And this is how mediocrity reproduces itself.<\/p>\n\n\n\n

The same is true for most websites. Perhaps 5 percent are genuinely bad and another 5 percent are excellent. The remaining 90 percent are simply average.<\/p>\n\n\n\n

\"\"\n\t\t\t\n\t\t\t\t\n\t\t\t<\/svg>\n\t\t<\/button><\/figure>\n\n\n\n
<\/div>\n\n\n\n

Mediocrity Reproduces Itself: Four randomly selected Tech Websites<\/strong><\/p>\n\n\n\n

<\/div>\n\n\n\n

Still, the AI industry wants us to believe that the right tools will allow us to achieve anything and turn every beginner into an expert.<\/p>\n\n\n\n

Every month brings another model, another framework, or another groundbreaking update that promises to increase our productivity once again. But the opposite is often happening.<\/p>\n\n\n\n

Optimization is starting to consume itself.<\/h4>\n\n\n\n

Anyone who tries to stay completely up to date at this speed will have very little time left to find genuinely good solutions. Many of us now spend more time testing new tools, learning how they work, watching video tutorials, and reading newsletters, posts, and articles than actually producing anything.<\/p>\n\n\n\n

For many self-proclaimed \u00abAI experts\u00bb the working day already looks something like this: 50 percent maintenance, 30 percent administration, and the remaining 20 percent spent producing interchangeable, average solutions.<\/p>\n\n\n\n

Eventually, we may realize that we do not need the latest model or the best tool to create outstanding work. More often, we need the right mindset, a little passion and energy, and above all, time. But that\u2019s precisely what\u2019s now being systematically taken away from us.<\/p>\n","protected":false},"excerpt":{"rendered":"

Claude can now \u00abdesign.\u00bb So naturally, I tried it.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"inline_featured_image":false,"footnotes":""},"categories":[39],"tags":[],"class_list":["post-1742","post","type-post","status-publish","format-standard","hentry","category-artificial-intelligence"],"_links":{"self":[{"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/posts\/1742","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/comments?post=1742"}],"version-history":[{"count":17,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/posts\/1742\/revisions"}],"predecessor-version":[{"id":1770,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/posts\/1742\/revisions\/1770"}],"wp:attachment":[{"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/media?parent=1742"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/categories?post=1742"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/thomas-sokolowski.com\/en\/wp-json\/wp\/v2\/tags?post=1742"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}