AI made me miserable

Recently I have been struggling a bit with both personal projects and also projects I have been taking on at work. Building and working on complex things, tinkering with projects at home, it just didn’t give me the same satisfaction it used to. Everything left me kind of empty, with no feeling of accomplishment or learning at the end of the day. Yet I was constantly under the pressure of that feeling of missing out, of not accomplishing enough.

This is a feeling I want to talk about today, and my ways of dealing with it. This post will be about my current experience and thoughts about AI (especially generative AI & LLMs).

AI force feeding

Let me start off by saying that this post is not about shaming all of AI and every user who is using AI tools. When used right, AI can have amazing use cases, both for society as a whole (think medical for example) and also as a tool for us individuals. Unfortunately, this is exactly where my current issues start: “When used right”.

In my opinion, most people (and companies) don’t use AI the right way. They use it to create soulless and meaningless content, to shove it in your face anywhere and anytime. While writing, I came up with the term AI force feeding, which I think is quite fitting actually. No matter where you are and what you do, you get more and more of this generated stuff pushed at you.

I think one of the worst examples of this is LinkedIn. Countless posts written with countless emojis (don’t get me wrong, I like emojis, but they are NOT a replacement for bullet lists), paired with an AI-generated infographic which is just as meaningless as the million other AI posts with infographics before it. I am convinced there is absolutely nothing genuine about these posts, with no human touch or authenticity. It’s just a wall of slop you cannot escape from. Pair this with using LLMs to answer your messages and comments for “engagement”, and you got yourself a perfect LinkedIn influencer in the making.

To be clear, the same thing happens everywhere, not just on LinkedIn. YouTube shoving their auto-translate down your throat without any option of turning it off (even translating titles and thumbnails now). Atlassian holding Rovo in your face whenever you try to use their products. Notion telling you that AI agents and AI meeting notes are what you need every time you just want to write down some thoughts. Developers generating walls of text and calling it “documentation” when they didn’t even bother to read it themselves before committing (and I have been there too, I won’t lie).

It’s everywhere, it’s getting worse and worse every day, and there is no escape from it.

To me it feels like we’ve gotten to the point where people will use AI for anything and everything, completely forgetting why they even do things. Not using their human sense to question if what they're doing has any sense or purpose. So many things are being done just because we can now, not because it makes sense. Come on, it’s not that hard to focus on a meeting and note down the five bullet points important to you. Nobody needs the 50th AI transcript of the meeting. Turn off that stupid transcript, focus, and enjoy a genuine conversation with someone. I think many people are genuinely not capable of doing so anymore.

More, better, greater

Apart from people just producing noise without anything to show for it, there are also many people out there actually building stuff with AI. Some things are more useful, many are less useful, but they all are projects people passionately built. And of course, people share all the stuff they build somewhere on the internet. To be clear, I am not saying they shouldn’t - I do the same thing after all. And the fact that people are sharing their creations is what makes the internet the (mostly) great place it is.

This quickly creates a false impression that everyone else is building cool stuff all day every day. Everyone else is getting rich right now. Everyone else but me. I think this triggers similar thoughts as social media. Instead of enjoying our own lives and being proud of them, we watch other people’s highlight reels and wonder why our own life seems boing and dull in comparison. Only with AI and “building” we (or at least I) compare the stuff I build to all the other stuff on the internet, all the highlights, and can’t help but feel inadequate.

My personal issue with this (and this 100% is a personal issue) is the fact that this triggers an infinite loop of comparison. Whenever I see a post somewhere (Reddit is my main source for this) that someone built something cool, it instantly triggers the thought of comparison in my head. “Why is this person building cool stuff, and not me?” “Why can’t I build something like this?” “Why am I wasting my time scrolling instead of building too?”

Feeling stuck

This ever-present comparison caused me to constantly feel stuck between the “pressure” of being more productive, building and doing more, and the feeling of exhaustion. Every day after work, my mind would tell me to sit down, be productive, build or do something cool. But just the thought of doing so left me feeling drained and exhausted. And a similar thing happened with “non-productive” things like gaming too. I just couldn’t sit down and enjoy time off, because my brain would constantly scream at me to “do and build cool stuff”. It was a endless cycle.

I know it’s unrealistic to be productive all day every day, to create and build all the time if I think about it rationally. But this did not matter to my head. We are all emotional machines at the end of the day.

Stacking on top of this was the fact that I was using AI tools (especially Claude Code) in a way which was not working for me. The output was there and I was building plenty, both at work and on personal projects. But doing so left me really drained and unfulfilled.

I think the feeling of exhaustion and being drained was caused by context switching. The way I worked with Claude Code was a loop of this:

1. Work with agent A to plan and build a feature
2. Switch over to agent B to plan and build another feature
3. Do mindless stuff (e.g. scrolling Reddit or LinkedIn) while waiting for the agents
4. Repeat

Another example (especially during working hours) was this:

1. Join meeting and listen
2. Claude needs something in the background
3. Context switch to what Claude is doing and review its output (while trying to somewhat keep up with the meeting)
4. Answer Claude, lose focus on the meeting
5. Repeat

This constant switching of tasks and context left my brain feeling like it was cooked in a microwave. Headaches and the inability to think straight were the result more often than not after these sessions. And especially not properly listening and being present during meetings feels unfair and disrespectful when looking back.

At the same time, the missing sense of fulfillment and accomplishment was caused by the fact that I didn’t really think while using Claude. At least not in a creative, problem-solving kind of way. Especially for personal projects, I was mostly just sloppily throwing prompts into Claude, loosely skimming over its plans and outputs, waiting to see if what it produced actually worked or not. And I think this is actually what was keeping me going for the longest time. That sense of “gambling” to see if the output the AI produced worked and did what I wanted it to, or if all it produced were more errors and garbage. I was okay with doing this with my personal projects because it didn’t really matter if something broke, but I realized that it left me drained and empty after, not satisfied with my progress and with no sense of accomplishment.

One article I read which actually captures my feelings and struggles quite well is Steve Yegge’s “The AI Vampire” - I highly encourage you to read it: https://steve-yegge.medium.com/the-ai-vampire-eda6e4f07163

Trying to catch myself

Recently I spent some thoughtful and also uncomfortable hours trying to understand my behavior and my feelings. I want to share what I found out about myself here, maybe it helps one of you out there too.

1. Thinking instead of consuming: I focus on doing the thinking, the conceptual and design work myself again. Sure, I use Claude as a sparring partner. But I don’t let Claude come up with the entire architecture and design of an app for example. I think of it myself, and only use Claude to do the “monkeywork” - writing down the Code with focused implementations and changes. I try to not let Claude dictate my thinking.
2. Learning and personal progress over constant speed and efficiency: Many of the issues I have been facing have been fueled by the constant need for speed and producing more. When reflecting on this, I realized that not the amount of output is what makes me happy and what I enjoy when tinkering and building. It is the feeling of learning something new and a sense of accomplishment and pride over the things I did and their quality. Looking back, I absolutely notice a drop in the quality and amount of personal learning I did as I started using AI tools more and more.
3. Avoid context switching: Instead of running multiple agents at once, constantly switching between those and other apps, I focus on one task or agent. Doing one thing properly absolutely beats doing three things at the same time with low effort. I actively noticed that my attention span and ability to focus decreased as I began to use Claude and AI more (which is worrisome on its own), and it has been hard to get better at this again.
4. Reduce socials to avoid comparison: If I am honest, many of these issues have been triggered by my constant comparison to others. This has been fueled massively by platforms like Reddit, which can be a highlight reel of “I built this” and “Here is what I did” by others. By cutting back on these platforms and consuming them more cautiously, I want to work against this constant comparison and sense of overwhelm.

Wrapping up

I want to wrap this post up by emphasizing that I am not dooming AI as a whole and I am not calling the technology evil. There are so many great things about it, allowing our society to advance and improve. But in my personal experience, I absolutely noticed that caution is required, or you get quickly sucked into some unhealthy habits. And I’d absolutely encourage everyone to take a step back and question their own (and their organization’s) use of AI and think twice if what they are doing is actually worth the toll and is actually producing any meaningful value in the end. And remember, there is no shame if you don’t max out your Claude subscription at the end of the week. Let AI work for you, not the other way around!

After noticing the patterns I described above and trying to apply my fixes as best I can I notice a steady increase in my focus and also happiness, while also feeling less exhausted overall. Its hard to constantly remind myself of these new habits I want to build, and I absolutely notice myself still drifting off, but I absolutely noticed and still notice improvements in my personal wellbeing and experience.

I am curious to hear your thoughts and experience on these topics! Feel free to leave a comment below! ✍️

If you liked what you’ve read, please consider sharing this post. Thanks!

Take care of yourself, and don’t let our AI overlords take control of you! Until next time 👋

Further reading

• Steve Yegge’s “The AI Vampire”: https://steve-yegge.medium.com/the-ai-vampire-eda6e4f07163