Welcome to the 367 newly Not Boring people who have joined us since last week! If you haven’t subscribed, join 233,187 smart, curious folks by subscribing here:

Today’s Not Boring is brought to you by… Neo.Tax

AI is good at some things and terrible at others. What the founding team at Neo.Tax — a Stanford ML PhD , a former IRS agent, and an ex-Intuit PM — understands is that AI is exceptional at digesting data within a defined ruleset. In other words, AI is great for taxes. 

Neo.Tax automates both the R&D tax credit and software capitalization (ASC 350-40) by connecting to your project management system, payroll and general ledger. Then, Neo.Tax produces an audit-ready study with task-level substantiation. 

The teams that know AI and finance, like those at OpenAI and Mercury, are already using Neo.Tax to automate away the pain for their engineering, accounting, and finance teams. No more surveys, interviews, or back-and-forths with your auditor.

Neo.Tax’s AI models, trained on the IRS codes, regulations, and tax court cases, act as your personal army of R&D tax specialists. Companies typically claim a 10% larger credit while saving hundreds of hours without compromising on accuracy (in fact, the IRS prefers contemporaneous documentation). You can even look back to clean up prior years or claim multi-year credits.

Use Neo.Tax for a smooth Q4 close or start fresh in the new year. If you’re scrambling for the October 15th extended filing deadline, they’ve got you covered, too. With year-round software capitalization, your R&D credit will be ready when you are—no hassle, no extra work.

Let AI do the heavy lifting, while you take the credit. Literally.

Check Out Neo.Tax

Hi friends 👋 ,

Happy Tuesday!

Dev turned four last Friday. His birthdays always seem to make me want to write something about raising kids today. When he turned 1, I wrote Dad Life. When he turned 2, I wrote How Do I Teach These Kids. When he turned 3, I wrote Riskophilia.

For his fourth birthday, I’m writing about living off-script.

Let’s get to it.

Burn the Playbooks

This morning, I saw a tweet that said “the modern world seems to crush geniuses.” 

Last week, I read that elite college students can’t (or won’t) read books.

The week before that, a startup that recently got accepted to Y Combinator with an open source AI code editor got roasted for ripping off another YC team with an open source AI code editor. 

These are all symptoms of the same disease: success has become overplaybooked.

Playbooks give you the rules of the game and the steps you must follow to win. They shrink the rich messiness of life down to an algorithm and turn people into machines. 

Ryan Yang-Liu’s argument – based on his own experience – is that math competitions take naturally very smart kids and focus them entirely on training for specific math tests. The game selects for people who are smart enough and can train the most and learn the best. In the process, it kills polymaths. 

The problem isn’t that kids are competing or studying hard, but that they’re narrowing. 

One of the reasons kids can’t (or won’t) read full books any more might be that books contain a lot of superfluous information that won’t be on the test. Mike Szkolka, a teacher quoted in the Atlantic article, explained that “There's no testing skill that can be related to … Can you sit down and read Tolstoy? ”, so teachers don’t make kids sit down and read Tolstoy. Or anything long and challenging. 

The problem isn’t that they don’t know Tolstoy, specifically, but that they might be missing out on the one book that ignites something in them that makes them want to learn.

My friend Mike loves reading just for the sake of it more than anyone I know, because of The Agony and the Ecstasy, a 1987 biographical novel about Michelangelo that his AP European History teacher assigned as summer reading heading into 10th grade. “It hit me at the right time,” he told me, “altered the intellectual trajectory of my life.” Three or four years ago, Mike ran into that AP European History teacher at a wedding. He told him how big an impact the book had made on him and asked if he still assigned it. “Oh goodness no,” he replied, as if the question itself were preposterous. “They could never handle it.” 

This carries into adult life, naturally, because how could it not? At a time when kids’ brains are most plastic, most able to soak in new ideas and make weird connections, we feed them playbooks. Then, they look for playbooks – for meaningless games to win – everywhere. 

The internet only figured out that that YC company was a fork of a fork after viscerally reacting to the cringiness of the original tweet. The tweet itself is formulaic: quitting the big job, giving up the big salary (always include big numbers), AI something something, 🧵⬇️, picture with the YC logo. And the fact that the kind of company that is so formulaic in its tweet – so devoid of creativity and originality – and in the product itself – a fork of a fork, so devoid of creativity and originality – got into YC hints that YC has become a game more than a signal. 

There is a pattern that repeats itself over and over. Maybe it’s always been this way. I wasn’t alive always, so I don’t know. I know what it looks like today: 

1. Someone does something for love of the game or out of pure curiosity. 

2. They have success doing that thing. 

3. Others see the success, deconstruct how they did it. 

4. Others still repeat those same things, joylessly, without the original spark. 

5. Create a cheap shadow of the original. 

I see this every day, in my particular line of work, every time a new startup hits my inbox promising to build X for Y but much better at Z in a $B market. Playing startup. Startup Mad Libs®. And I feel bad because you’re supposed to applaud the heroic effort required to start a startup, but a lot of these don’t feel heroic. They feel safe. And they feel doomed by that safety. 

I bet you see it too, in yours. 

Nothing great was ever created from a playbook. 

Playbooks are the death of creativity and joy.

Sure, refer to the playbooks for non-core things. 

If you’ve used all of your creative powers to come up with a cure for cancer and you want to follow a playbook for the FDA approval process, by all means, do that. 

But it seems as if too often, people are going through the FDA approval process in hopes that it will lead them to the cure, metaphorically speaking. 

They make the playbook the main thing. The playbook should never be the main thing. Playbooks are too easy to copy. 

The sad part of the three examples I gave – math tests, reading, and copying code – is that, very soon, AI will be able to do those things better than any human can. It kind of already can. 

Announcing its newest model, o1, OpenAI wrote, “OpenAI o1 ranks in the 89th percentile on competitive programming questions (Codeforces), places among the top 500 students in the US in a qualifier for the USA Math Olympiad (AIME), and exceeds human PhD-level accuracy on a benchmark of physics, biology, and chemistry problems (GPQA).”

I don’t think AI will replace humans. My bet is that we’ll end up in a Goldilocks Zone, where it gets really good at doing the things it does without encroaching on the creativity that is humans’ greatest resource. 

AI won’t replace humans, or beat humans at what we do best, unless we turn ourselves into cheap AIs. 

AI will do math tests better than we can. 

AI will do reading comprehension tests better than we can. 

AI will copy code better than we can. 

AI will follow playbooks better than we can. 

Anything that we playbook becomes more legible to AI. 

So we must burn the playbooks.

This is the real Butlerian Jihad, not destroying the machines, but destroying the playbooks that would turn us into machines. 

The challenge with an essay like this – one in which I’ve so vehemently railed against playbooks – is that I can’t very well turn around and tell you exactly how to do that. That would be a playbook. You need to figure it out for yourself. 

Maybe you want to read full, obscure books. Maybe you don’t. Maybe you want to do math problems for fun, or code for fun, or build something for fun, or call in sick tomorrow to hike in the woods, or spend the next 730 days straight working 18 hour days on the thing that you need to create because the world needs it and you can’t stop thinking about it and even though it will probably fail you’re willing to sacrifice years of your life for it, miss birthday parties and weddings and casual hangouts with your friends just to bring it into the world. Maybe you want to spend more time with your kids and family and friends. 

My own personal rebellion has just been to spend less time scrolling Twitter – maybe the clearest manifestation of how things go wrong when people bend themselves to please an algorithm – and more time reading good books about anything that interests me. It’s been wonderful. I think it’s made me better at my job, too. 

But I certainly don’t have all the answers, even for myself, let alone for you. Do something that makes you feel more creative, more alive, more human. Something that sets your brain on fire. Something that you love doing more than anything else in the world. Something that no one can do just like you. Something that can’t be playbooked.

You are not a machine. Don’t act like one.

Thanks to Claude for editing.

That’s all for today. We’ll be back in your inbox on Friday with a Weekly Dose.

Thanks for reading,

Packy