Hyperfocus Is Your Superpower In AI Slop Times
Code is not the bottleneck, noise is.
The cost of mediocre automatically generated content has dropped to zero. Over 70% of the things my parents and older relatives share on our group chats are AI slop. I would have assumed this is a problem for people who are not in tech. But many of the inbound messages I get are poorly written with an LLM.
Blindly Trusting LLM Outputs Is Sloppy Work
I agree with the conclusion. It should have career consequences just as any type of half assed work:
“Sorry for the vent but I just spent 30 min reviewing a colleague’s documentation full of amazing user quotes in support of a feature. Which turned out to be mostly misinterpreted, misquoted, or even just completely hallucinated (as in, quoting from a person who doesn’t exist, or a quote between quotation marks that is not in the referenced doc).
I feel like this needs saying: professionals don’t just make shit up, and using an LLM to help is no excuse. This is not a dumb junior getting caught copying on the test for the first time. This is someone with a reputable 10+ year work history at a number of companies.
The problem with LLM is that the cost of generating bullshit now vastly undercuts the cost of detecting and correcting it. I probably spent more time following links and verifying sources than the author did asking some bot to put together the reference section of their doc. As with similar situations, I feel this calls for punitive consequences.” - source: OpeningBang user on Reddit
Figuring Out What To Build Is Still Hard, For Now
Adding to the above an example of what remains the core of knowledge work and not necessarily just for Product Managers:
“But the actual hard part of my job - figuring out what we should build based on contradictory customer feedback - AI hasn’t touched that. I tried feeding it customer conversations and asking for patterns. The output looks impressive but misses the nuance that matters. Like, it’ll tell you “users want better reporting” when the real insight buried in those conversations is that they don’t trust the existing data and the reporting UI isn’t the problem. - source: frustrated_pm26 on Reddit
Hyperfocus Is Your Value Add In Bussiness … As Well As In Life
I was on a panel about AI with on 05.02.2026 in Madrid. The event, hosted by Product People was on AI-Powered Discovery: How to Learn From Users in Hours, Not Weeks.
Prashant Kumar, one of the panelists stated that hyperfocus is a must for product leaders and executives since AI expands options and also creates a lot of noise. That rings true since code is no longer the bottleneck thanks to Claude Code, ChatGPT Codex, Cursor, Windsurf and others. Product Discovery, positioning and go to market speed are currently more valuable.
I’ve been working in tech for and consulting for a while and I often interact with highly productive and competent folks that show neurodivergent traits (which also correlate with high intelligence). Didn’t test this hypothesis yet, so I could be wrong. For now, I’m of the opinion that it’s even a plus in this day and age as it “protects“ people from the traps that draw everyone else in for now. I’ve observed these people are being able to hunker down and work for very long on a few narrow topics, compartmentalize well, optimize their personal routines and even information diets. Have tried some tactics and will do a post of things that worked and didn’t for me.
“Humanity” As A Premium Product
If the public web and advertising driven social media platforms are filled with slop and bots, where does that leave the real people?
In 2026 this seems to be in interest based communities with chronological feeds. (Reddit, Telegram, Discord, WhatsApp Groups).
Stats:
Reddit’s Daily Active Uniques (“DAUq”) increased 19% year-over-year to 121.4 million in 2025
Telegram surpassed 1 Billion MAU (Monthly Active Users) in 2025 according to the Founder, Pavel Durov
Discord
Personal Predictions for 2030
Let’s see if these age like wine 🍷 or like cilantro 🌿
We’ll fully embrace AI content and won’t even know the difference as it gets get better and better at farming social media platforms. “If you can’t tell, does it matter?“ the question a robot asks a human in Westworld S1E1.
Hikikomori (people who isolate themselves and typically don’t leave their home) will grow 3x across the world driven by AI Companions, radical opinion divides on social media platforms and cost of living. It’s currently at 1.5-2.5% (those who do not leave their house for 6+ months) and 8% (when using broader screening tools like the HQ-25)
In person meetings will be the only “real“ human thing left. Hanging out without devices feels already like the highest form of appreciating someone and showing up for them. Commercial areas that facilitate this will charge a premium.
Asking yourself everyday 'What is one truth you know that everyone else will disagree with?' (from Peter Thiel's 0 to 1 Book) was and is still is the best way to practice your mind muscle. Contrarian views will raise ears when mediocrity is around which is often the views that most people agree with.