In today’s fast-paced world, many of us find ourselves grappling with overwhelming choices and pressures, especially when it comes to technology. Let’s delve into a prominent conversation around productivity, AI, and the quest for balance.
Where do I even begin? It appears that many individuals are experiencing a similar struggle. The continuous influx of new products, models, and capabilities can be overwhelming. It’s particularly disconcerting to witness others managing their lives with multiple personal AI agents.
What is our goal in this chaotic environment?
When I initiated the newsletter, it was a means to stay abreast of developments happening in the industry. Initially, it was five times a week, but I had to scale it down (along with our revenue) to just two days due to the overwhelming noise. Although I strive to share valuable links, the clutter persists, and it might even be louder now.
Recently, I came across two insightful posts that echoed these sentiments. Here’s a brief excerpt from each:
You’re doing your best to keep up. You’re working 9-9-6, sometimes 0-0-7. You didn’t actually rest during your supposed time off, did you? What tasks are your agents managing while you’re reading this? Have you tried the latest personal assistant? Have you optimized your workflows to suit the latest improvements? Have you experienced viral success on X recently? Have you engaged in angel investing? Have you locked in your Bay Area mortgage? If not, you’re in deep trouble, right? You’re tethered to your keyboard, timeline, and Slack. You’re fatigued, yet your ambition feels like Icarus soaring too close to the sun. Your fear of missing out is soaring. Your capacity is nearing its limit. Is your moral compass spinning?
The Sane Asylum by Colossus
To be candid, instead of allocating time to “do” a task, you can remain on your couch, hungover and in sweats, fire up Claude Code, Codex, GrokBot, or any similar tool, and dictate a new project. Not feeling like typing? Just double-tap the function key and space bar, instruct Wispr Flow to transcribe, and say, “I want you to develop a tool that summarizes key highlights from all the newsletters I subscribe to and delivers this summary daily.” Then go back to enjoying your downtime while your AI does the grueling work.
The Difference Between “Can” and “Should” by Jack Raines
I tend to initiate more projects than I finish. I have a course that’s about 70% complete, and I’ve built a tool for non-techies using personal agents that’s approximately 80% done. Furthermore, I still need to finish fundraising for an outstanding fund. I should pitch a new sponsorship model to various companies.
While I should pursue tasks that make business sense, my motivation often stems from the enjoyment of the process itself. I only find the drive to engage in something if it feels fun.
And therein lies an issue for many people: play seems to be absent. We’re often driven by our employers or peers to accomplish more. But to what end? It becomes busywork—cleaning the desk before doing the real work. AI was designed to alleviate our workload, yet it often highlights the endless tasks we could undertake.
Just one more attempt at the slot machine.
There’s no need to engage with every tool available. You don’t require a myriad of agents optimizing your existence. I personally dislike optimizing anything.
I recently unsubscribed over 60,000 individuals from this newsletter because they weren’t actively engaging with it. While those numbers look good for sponsors, that’s not the purpose of this platform. I want to share what I find interesting, what I do for fun. I tinker and create spontaneously. My goal is to demonstrate how one can utilize AI creatively without succumbing to a hyper-productive mentality.
The videos I made turned into a quest to optimize them. Am I aiming to become a YouTuber? Certainly not. If you don’t enjoy the videos, that’s perfectly fine; you’re not obligated to watch them. I’ll create them when it feels right, focusing on quality.
I find myself heavily preoccupied with these reflections (as you might have gathered). I was once a proponent of not needing coding skills when everyone claimed it was essential. Yet now, I find myself entangled in managing multiple agents to optimize everything, believing it would boost my productivity.
In a business context, hiring is often justified only when a pain point can be resolved through additional personnel. So why am I bringing on (and managing) various agents for tasks that aren’t significantly troublesome?
Most individuals can realistically handle about four concepts at a time. Admittedly, that may even stretch my capacities, but it holds true. Top-tier engineers can produce substantial output and multitask, making those the organizations I wish to invest in, as they might be what drives the next generational trillion-dollar company. However, for those seeking to move swiftly yet deliberately, we must identify alternative routes.
I want to emphasize the importance of fun. AI offers me capabilities I never imagined possible, but those pursuits do not always need to be valuable or relentless. If I’m passionate about a task, naturally, I’ll employ an agent to assist. However, it should not work for me; it’s just a tool, not a person to manage or worry about.
I may not have all the answers, but in the upcoming 30 days, I plan to undertake several initiatives:
-
Engaging in ‘morning pages’—first thing every day, I’ll write down my thoughts to clear my mind.
-
Utilizing agents only for projects that genuinely interest and excite me.
-
Avoiding new tools unless they clearly address a personal challenge.
-
Skipping lengthy post summaries and choosing to read them or move on.
-
Not doing anything solely for monetary reasons.
I’m already thinking of ways to package a concept around “moving fast and slow with AI” into a business framework. However, I will consciously invest more time pondering it before abandoning it to an agent. I aim to experiment with videos, writing, and interactive content that energizes me.
I didn’t intend for this to be so lengthy, but here we are.
If any of this resonated with you or if you have thoughts to share, please respond—I’d love to hear your perspective!
I must give a shoutout to Ashe, who messaged me after my tweet, and we had a fascinating phone call. I have great admiration for her boldness in pursuing her vision, and she provided me with some valuable insights to reflect on in my journey.
-
Google has rolled out two new models:
-
Gemini 4 Carbon. A significant improvement over Argon in benchmarks, though currently only accessible to cybersecurity defenders.
-
Genie 4. A world model capable of creating and editing 3D scenes (like changing weather or incorporating characters).
-
-
Suhail’s new venture, Particle, is focused on training coding models to achieve results with fewer tokens.
-
Four personal agents are tasked with a nine-task test. Which one actually gets things done?
-
A free, open-source guide to assembling IKEA furniture, complete with 3D instructions and voice guidance.
-
How agents influence experts – it turns out 86% of the time, experts follow an agent’s arbitrary suggestion.
-
Simon Willison argues we’ve reached an inflection point for vibe-coded games. They’re genuinely entertaining now.
-
Readwise Books – a new reading and tracking app that integrates with your agents.
-
Clicky has begun to generate custom visuals on your screen when you ask it questions.
In conclusion, as we navigate this complex landscape filled with technologies and myriad productivity tools, it’s crucial to find joy in our pursuits. By focusing on enjoyment rather than obligation, we can strike a balance and ensure that our work remains fulfilling and meaningful.