Approved Thoughts

I have been following the AI revolution almost since the day ChatGPT launched in Nov 2022. That this was a transformational technology has also been clear to me for almost as long. I even sensed the "vibe shift" in early Jan.

But so far I didn't really "feel the AGI". Last week I did.

AGI stands for Artificial General Intelligence. Here's how Claude defines it:

An AI system capable of performing any intellectual task that a human can — reasoning, learning, and adapting across domains without being specifically programmed for each one." - Claude Sonnet 4.6

It is also worth knowing the related concept of ASI, Artificial Superintelligence.

An AI that surpasses the cognitive abilities of all humans combined across every domain, including scientific reasoning, social intelligence, and creative problem-solving."

Last week we finally signed up for Claude Code at work and I started playing with it.

Claude Code works via the CLI (command line interface) or the terminal - the black screen with white text used by all the movie nerds. It can be a little intimidating if you are not a programmer but really once you set up Claude Code, it works just like the chat window.

It is so much more powerful though. It can manipulate files on your computer and run the code it writes. This means it is not restricted to recommendations or single steps any more. It can generate an entire plan of action and execute and implement the thing by itself. It can be more than a little freaky when you see the output of a particularly complex task.

Let me share a couple of examples that blew my mind.

At work we have a database that stores our financial projections for a company we cover. The investment team can use custom functions in excel to then download this data. This is useful to create reports for analysis - say comparing 5 companies across a few metrics. There are different functions for different types of data and the IT team has created an excel file with about 20 sheets listing the syntax for each function, the list of values that can go in each function etc.

I pointed Claude Code to this help file and asked it to create a "skill" for itself that would allow it to create reports in excel using these formulas. I gave it the same context as the previous paragraph - maybe a little more technical but nothing a lay person wouldn't understand.

With that single command, in maybe 3-5 minutes, the skill was ready. Now when we need to create a report we can just ask Claude Code to use the skill, the data we want in the report and it creates a fully formatted excel file with the (usually) correct formulas. Tasks that would take me or my team members 20-60 minutes, automated permanently. Using English sentences, no technical knowledge.

Second example. We wanted to perform some statistical analysis on 20 year historical performance of 600 stocks to identify specific episodes / time periods and then dig deeper into specific episodes to understand their fundamental causes.

An analyst spent probably 5-6 hours to download the data in excel, process it so we could start identifying the qualifying episodes / time periods in different markets. At this point we were somewhat stumped about how to isolate the relevant episodes from this vast data. Probably a trivial problem for a data scientist but not for us.

In the past, we would have spent probably another 5-10 hours trying to either eyeball the data using charts or some other way to get our answer. Over the last couple of years, we would have asked Claude or ChatGPT to suggest a better way.

But since we had Claude Code, I pointed it to the existing file, with all its messy sheets and structure and explained what we were trying to do and asked it to identify the episodes. It whirred away for about 20 min, an occasional question here or a permission there and then it spat out a report.

But the report didn't just have the episodes identified. It also identified potential causes for each episode (based on web searches presumably), linked patterns across multiple episodes, gave charts and commentary that helped us understand the relevance of each episode, limitations of the analysis, suggested next steps etc.

Now we have all seen more than enough AI slop to not be impressed by the sheer volume of content these tools can spit out. But we spent a few hours verifying the numbers and conclusions. So far it all checks out. You have to take my word for it, but man, this was not slop. If a junior associate had put this out I would be proud of them. There were parts that I would have been proud to create and, of course, large parts that we simply couldn't have created at all.

And those were not the only thing Claude Code did last week that blew my mind.

So, let it be noted on this 8th of March, 2026, I felt the AGI last week. I may even have felt the ASI.

The big news in markets this week was this report from Citrini Research & Alap Shah that apparently crashed the markets and led to a lot of debate in our office. It lays out a "fast take-off" scenario for AI, which causes mass layoffs of white-collar emplopyees as AI replaces intelligence work and starts off an economic downward spiral as demand collapses.

It should have been clear all along that a single GPU cluster in North Dakota generating the output previously attributed to 10,000 white-collar workers in midtown Manhattan is more economic pandemic than economic panacea. The velocity of money flatlined. The human-centric consumer economy, 70% of GDP at the time, withered. We probably could have figured this out sooner if we just asked how much money machines spend on discretionary goods. (Hint: it's zero.)

AI capabilities improved, companies needed fewer workers, white collar layoffs increased, displaced workers spent less, margin pressure pushed firms to invest more in AI, AI capabilities improved…

It was a negative feedback loop with no natural brake. The human intelligence displacement spiral. White-collar workers saw their earnings power (and, rationally, their spending) structurally impaired. Their incomes were the bedrock of the $13 trillion mortgage market - forcing underwriters to reassess whether prime mortgages are still money good.

The report found many believers in markets but I find myself on the skeptical side, much more pusuaded by the many pushback articles which are grounded in conventional economic theory. And they came from many sources.

Here's Tyler Cowen in his cryptic style. Here's Zvi with the inverse. And finally here's Citadel.

And lastly, here's Claude summarizing it all and adding its own perspective:

Why Citrini's Scenario Doesn't Add Up The piece is an excellent thought experiment and a useful sector-level vulnerability map. The macro conclusion — that AI abundance causes a demand collapse and systemic crisis — is built on a fundamental accounting error.

The Core Contradiction: Every Loss Is Someone Else's Gain The entire scenario rests on a demand collapse: AI replaces workers, workers stop spending, the economy spirals. But the same force destroying jobs is also destroying prices. If a Claude agent does the work of a $180K PM for $200/month, then everything that PM helped produce also gets dramatically cheaper. The piece catalogs agents slashing insurance premiums, SaaS costs, delivery fees, real estate commissions, and interchange — then claims displaced workers can't afford things. Which things? The things that just got 80% cheaper?

Every corporate revenue loss in the piece is a gain on the other side. ServiceNow loses $500K in licenses — that's $500K freed for the client. DoorDash loses its 30% take rate — drivers earn more, consumers pay less. Real estate commissions drop from 6% to 1% — that's a 5% stimulus to every home purchase. SaaS fees are a tax on business. That tax went down.

Meanwhile, the piece describes NVIDIA posting records, hyperscalers spending $150-200B/quarter, AI companies thriving. Someone is paying for all of that. You cannot have booming AI revenues and an economy where nobody is spending. The money doesn't vanish — it circulates through different channels. The piece tracks one side of every transaction and ignores the other.

The Timeline Is Physically Impossible The scenario requires mass deployment across every major industry in ~30 months. Enterprise procurement cycles, organizational restructuring, regulatory review, and infrastructure buildout don't move at this speed. More critically: there isn't enough compute on Earth to run always-on agents for every American consumer while simultaneously replacing most white-collar work by 2028. The compute constraint alone paces everything and makes this timeline a non-starter. Even clearly superior technologies (electricity, the internet) took decades to restructure economies.

Zero Policy Response for Two Years Is Ahistorical The US deployed trillions within weeks during COVID. Automatic stabilizers activate immediately when unemployment rises. The piece treats political difficulty in designing optimal long-term policy as equivalent to no response at all.

More importantly: in a world with booming real output and deflation, aggressive fiscal stimulus is straightforwardly correct — you have too much output and not enough demand, so you print money without inflationary risk. During COVID, stimulus risked inflation because real production was down. Here, real production is exploding. This is the easiest macro policy environment imaginable.

What's Actually Useful in the Piece The sector-level analysis of which business models are vulnerable is genuinely good. SaaS pricing power, intermediation platforms, payments, insurance built on inertia, real estate brokerage — anything dependent on friction or information asymmetry is legitimately threatened. The private credit / insurance / reinsurance daisy chain is a real and underappreciated fragility.

The transition will be uneven. Specific sectors, geographies, and demographics will get hit hard, and the speed of disruption concentrated in high-earning white-collar cohorts could cause real dislocations even if the macro aggregates are fine. That's worth positioning around.

But the mechanism for how sectoral disruption becomes a macro demand collapse is where the analysis fails — because it treats every efficiency gain as pure destruction rather than a transfer from incumbents to consumers. Read it as a sector vulnerability map, not a macro forecast.

Going back to the title of my last post from Singapore, there's clearly been a vibe-shift when it comes to what the latest models can do. In early Jan only those focused on AI were feeling it and now it is going more mainstream. But that's no reason to fall prey to half-baked thinking. Let Claude taste it and cook it for you.

The Remains of the Day by Kazuo Ishiguro, and the Oscar-nominated film.

I recently read The Remains of the Day by Kazuo Ishiguro, after coming across a positive review and remembering that there was an Oscar nominated movie based on the book in the '90s, when watching the Oscars used to be in thing in India.

After the first couple of chapters, thinking that this was perhaps an artsy book, I also opened up a chat with ChatGPT to improve my understanding of the book.

This was useful but note to self: in future explicitly tell ChatGPT to avoid spoilers or otherwise reveal information or conclusions from the later chapters. To be fair, I did ask for themes to notice in rest of the book so its mostly on me.

This isn't a thriller so in one sense the damage was limited. It is, however, a very subtle book that gradually unveils its layers and getting a bunch of bullet points upfront, spelling out every complexity and nuance wasn't ideal.

The book is a masterpiece. Very few books have ever had this effect on me where, for days afterwards, I suddenly get a flash of feeling, a pang that forces me to stop for a breath.

The closest analogy that came spontaneously to my mind was of eating great Japanese food (yes, even as a vegetarian) - the flavours are extremely subdued - very much the opposite of, say, Indian food. But they are unmistakable, sharp even, if you stop and notice, and you know it took a lot of effort and craftsmanship to pull it off just right.

When reading fiction, I have a tendency to really flip pages towards the end, a tendency developed from years of reading mostly thriller novels. Because of this I found the book depressing, having barely spent any thought on the final pages. It is only while writing this post that I revisited the ending (and the title) and had to reconsider my understanding.

I had identified so deeply with the sense of loss in the middle of the book that I completely missed the gentle ray of hope that Ishiguro leaves us with at the end: as humans, on any given day and at any given time, all we can do is to make the best of whatever remains of the day.

Highly recommended.

As a viewer, I almost never think about the person behind the character. Sure if a Christian Bale loses nearly 30kg for a role, its visible enough and stark enough that one is forced to think about the how of it. But for the most part, I watch the program and move on. Maybe if there is something really interesting or shocking, a comment to my SO.

So it is a jarring to see that the impact of a brief scene stuck with the actor for years after the fact.

These days there is enough nudity on screen that most of us are desensitized to seeing it and I hope the actors are too, to performing them. Even in 2006 or 2012, when maybe it was less common, I would not have imagined an actor, and especially an American¹ actor, would be so traumatized by such a scene.

Maybe it is because she grew up in a time when that was a much bigger deal that it was harder on her. Perhaps actors like her were the trailblazers and paid a price and it is easier for actors now.

It is, however, possible to imagine a possibility where being naked on screen is traumatizing, no matter how common and acceptable. According to Claude, the "being naked in public" dream genre is common across cultures and age groups and more common in adoloscents.

If that is the case I shudder to imagine the cost of all the content on Netflix and other services in a few years time.

The report also has a bit on another scene she was not happy about. And guess what? At the time (and thankfully no longer), the report carried a photo from that scene.

I need to sit down and combine some of these thoughts into a coherent piece but for now, I just want to dump it all here for reference later. I doubt any of these are original and I have certainly seen some elsewhere.

1. Market structure for LLM makers may end up being like the airline market. High fixed cost to set up, hard to create a product differentiation that users care about (other than price), lots of competitors entering, at least in part due to the "prestige" of owning one. Lots of utility for consumers but hardly any profit for producers.

2. An LLM with an infinite context window, one that can contain all my life, will be an entirely different product than an LLM with a limited context window. You can never have enough when it comes to context windows.

3. Notwithstanding 1, "personality" makes a huge difference in the experience of working with an LLM and the ability to create the right one could determine whether a model can dominate a market or a niche. Claude Sonnet 3.5 absolutely had than special sauce in my experience. We need a lot more of it.

4. This.:

   I don’t see how we’re going to avoid a situation where the internet become lousy with AI-created, pseudo academic writing filled with made up facts and quotes, which will then get cemented into “knowledge” as those articles become the training fodder for future models.

   But combine it with the fact that if we all start getting our answers from LLMs, the online content & ad based business model goes caput and then what is the incentive for people to put up good content on the web? None.

5. But anyone who has good proprietary, verified, high-quality data & content will potentially control the value for the customer even as base LLMs become a commodity. Therefore does more data and content start going behind the paywall? If it doesn't it becomes training data and cannot be monetized.

This is the website in question. Bookmark it. Or don’t.

Many people are asking if this can even be called a website. These people are losers. Sad. The finest people I know think its a great website, maybe the best website in the history of the world. Nobody knows websites like I do.

Anyhoo.

It all began when I decided to shift this letter to a website last week. Having purchased a few web domains before, it took me only a few minutes to buy approvedthoughts.com from Cloudflare. Unlike GoDaddy etc. who offer a promotional price bundles and then jack it up on renewals, Cloudflare offers at-cost registration + renewals and I have good experience with some of their other services, so it was an easy choice.

For all my talk about using AI models, I hadn’t really used them for any type of programming project / tech support roles. I figured this was the perfect opportunity for me to try out if I could actually create something useful with the help of one of these models.

For context, while I have an undergraduate degree in Computer Applications, I have never programmed professionally and frankly did barely any programming during the course itself. I am probably a bit better than someone who has never done any programming, but not by a lot.

For context also, I could obviously just spend $20 on a paid service that would be absolutely trivial to set up and manage but where’s the fun in that?

This is how I started on Claude 3.5 Sonnet:

I have recently purchased a domain for a website. I used cloudflare for purchasing the domain. I would now like to set up a Wordpress or other website at the domain. Walk me through the steps of doing this.

The Sonnet is the “most intelligent model” by Anthropic. It is the go to model at the moment for a lot of the people I follow and the only model I pay for.

The project went through multiple chats because after a while, in each chat, I would start getting this message:

Here, for the first time, I used a trick I had recently seen (I think) on Twitter.

ok - I have saved and deployed on cloudflare. However I am getting the message that this chat has become too long. Can you provide a status update that I can use with another AI chatbot to continue this work?

And it did! I pasted the update in another chat and the conversation continued there as if it was the same chat!

You can check out the first chat [here. I think / hope it has no confidential information. I won’t be sharing the remaining chats because they do have some of my account credentials. This one should provide enough context for most of what I am discussing here though.

So How’d it Go?

Well you’ve seen the “website”, so clearly not as well as I’d hoped. That said, I am continuing with the project and based on what I have learned so far, I am hopeful I will eventually get it working.

It is still remarkable that I got this far, considering that I have zero experience with the technologies / services involved. Claude successfully walked me through setting up 3 different services (Cloudflare, GitHub and Hugo) and had them work with each other to serve up the site.

All of this happened within a couple of hours and with me blindly following its instructions - copying and pasting code where it asked me to, choosing the options it asked me to and only making high level decisions, like which theme I liked for the blog, which require no technical sophistication.

For the most part, all of this was “one-shot”, meaning it told me to do something, in plain English, I did the thing and it all worked. Where it required a second shot, it was because I hadn’t provided the full context e.g. that I was using a Mac and not a PC. And once I clarified that, it adjusted the directions for all subsequent steps.

In fact, because I use Claude almost exclusively, I had forgotten how much better it is that some of the other models at being almost “human”. This chat is not very different from how I would interact with IT support engineer at work. Just randomly dropping bits of context or asides and having it be incorporated correctly into subsequent interactions as relevant.

In one of the subsequent chats, it seemed to me like Claude was stuck or going around in circles, without fixing the problem, so I decided to switch to the newly launched DeepSeek-R1 reasoning model. While the R1 made immediate progress where Claude was stuck, the experience of working with the R1 was just so much harder.

Unlike Claude, which “understood” that I was not technically sophisticated, and only gave tasks of a manageable size at each stage, the R1 would just throw a laundry list of complex tasks at me.

For e.g. there’s a set of 3 commands I needed to run on the Terminal app on my Mac, after every change to the code. This would “push” the changes to GitHub. Every time there would be only minor changes to these commands. Claude would provide these commands every time, unprompted. R1 would not.

Claude also seemed to just understand better where we were in the overall project at all times. So if it asked me to do three things and if I went back with a question about the first thing, it would help me fix that and then go back to the remaining two items in the list. R1 did not seem to have as good an “understanding” of the overall project and did not seem to “appreciate” that if I was asking a question about step 2 of 5, that steps 3-5 still needed to be done after it answered the question about step 2.

Of course, the answer to step 2 would have 5 steps, so now I have to scroll back and forth to figure out which step of which step I am on. So despite being a “stronger” model, I found it discouraging and just much harder with the R1 to have to keep track of where we were in the project.

What went wrong?

I suspect the website would be looking a lot better if I had just followed Claude’s guidance. In fact, even now, I think it is at a stage where I can post these weekly updates there with some extra effort.

But I knew I was going to be traveling for a few weeks and so after the site was ready for me to start posting content, I asked Claude if we could also set up a CMS, content management system, for the website. That way I could post from anywhere instead of just from my desktop and include links and images in the posts with minimal effort.

Things went off the rails pretty much immediately. Without getting into a lot of detail, we encountered a bug while setting up the authorization setup for the CMS. Claude kept suggesting various tweaks but nothing worked. It felt to me like we were going around in circles, trying the same things over and over again.

I even asked Claude to look at the entire conversation and figure out if it was going around in circles. It gave itself a clean chit. Of the 6 or so hours I have spend on this project, my guess is 4-5 have been trying to resolve this issue.

My suspicion is that Claude is trained on an older version of the various services / projects I am using and therefore could not fix these errors.

Eventually I gave up with Claude and took the project over to DeepSeek R1. It made some progress, identifying some errors Claude had made, e.g. key files located in wrong folders, wrong parameters etc. However since neither of the models can really take over and look at the code themselves, they may not be able to fix issues caused by my mistakes in implementing their directions.

Maybe AI agents will fix this issue in future. In any event, the project is on pause as I travel and I am wondering whether I should start from scratch with DeepSeek instead of trying to fix the current version.

What did I learn?

A lot.

Despite the hiccup, I feel comfortable that these models can help me do projects that I just couldn’t have done before. I intend to try more projects, and of course, finish this one.

Different models have very different “skills” and it’s important to figure out which model is the best for a given task.

While Claude is the easiest to talk to, almost like a person, it really helps to understand how to prompt other models to get the best out of them. Providing more context and better instructions can result in much better results.

Larger context windows will change everything. A model that can keep your entire life in context will be multiple orders of magnitude more helpful than the current models.

I am a little more skeptical than before about LLMs becoming generally intelligent or superintelligent. Their ability to deal with problems that are not in their training data seems limited. Maybe agentic models will overcome some of this but breakthroughs in for e.g. fundamental physics seem unlikely. Obviously I am extrapolating from a tiny sample but that’s the direction in which I have updated my priors.

What’s next?

I have a few ideas I want to work on once (not if!) I finish the website. I wonder if I can set up a web server on my old Mac and run a few basic services off it. For e.g. creating a URL shortening service (like Bitly) or a QR code generation service.

This from Simon Willison is my inspiration. 14 projects in 1 week. No rocket science but just small, useful things here and there. Or, if I am more ambitious, that custom Spaced Repetition app built by Andy Matuschak, I mentioned couple of weeks ago.

Is truly customized software possible? I hope it is.

Having done a throat clearing piece the first week and the one writing idea I actually had when starting the newsletter, the second week, this week I was finally forced to think about the direction I want to take with this letter.

I enjoyed heading down the rabbit hole of how to learn better last week. I already knew that there was much I needed to learn about teaching after struggling to homeschool the boys for a year recently. Last week it also became clear that I also had to learn a lot about learning.

As it happens, this is also a space where a lot is changing (yes, AI, duh!) and one that is even otherwise relevant to me given the presence of two (reluctant) learners that I am (even more reluctantly) trying to support.

So it seemed like a natural beat to walk.

I subscribe to a few Substacks and as I trawled through them I realised that the only reason I subscribed to each of them was because they were written by experts who also happen to have an amazing way with the words. Here’s a sample:

1. Money Stuff by Matt Levine - GOAT of newsletters. It’s not even close. And daily? Get outta here…

2. Don't Worry About the Vase by Zvi Mowshowitz - I can’t believe it’s free.

3. Bits about Money by Patrick McKenzie (Patio11) - How is a software engineer teaching me about my job?

In other words, I have a very high bar for who I let into my inbox. I could write about how to learn better but clearly I am not the preeminent expert on the topic, I am barely getting started. It would be hard for me to clear my own bar for quite some time, if ever. And if I am going sustain this, I need to enjoy what I am doing. I need to be able to experiment, make messes and generally figure things out.

It occurred to me that a better medium for that kind of a goal might be a website rather than spamming people in their inboxes. It will give me the freedom to play around with the format, topics, content & timing and make the aforementioned messes without worrying too much.

Now my wife thinks I am trying to wiggle out of my New Year’s resolution with this talk, but frankly, she barely knows me, so we can safely ignore that possibility.

So here’s the plan: I will continue writing here on a weekly basis till the website is setup and then move everything over there.

What is Spaced Repetition (SR)?

Claude says:

Spaced repetition is a learning technique that involves reviewing information at gradually increasing intervals. Here's how it works:

When you first learn something new, you review it fairly soon (like the next day). If you remember it well, you wait a bit longer before the next review (maybe 3 days). If you still remember it well, you extend the interval further (perhaps a week, then two weeks, then a month).

The key principles behind spaced repetition are:

1. The spacing effect: Information is better remembered when studying is spread out over time rather than crammed into a single session.
2. The testing effect: Actively recalling information strengthens memory more effectively than passive review.

This may not sound revolutionary. After all, some version of this is how we prepared for our exams. But the important thing is that unlike the 90s, there’s an app for that (Anki, Mochi & others). The app handles the spacing and repetition, making it more practical, reliable and scalable.

When I first learned about and used SR apps years ago, I thought, “Wow! It would have been great to have this in school. This will be great for the boys in a few years.”

I never used the technique until last year when I used Anki for homeschooling the kids in Hindi. You can get many ready-made decks (free & paid) on the Anki website, but creating your own (necessary for specific requirements) was a chore, and I stopped after a few months.

I thought SR was for students until I saw a Dwarkesh Patel podcast with Dan Shipper using it to learn more broadly. It wasn’t limited to coursework or vocabulary but included concepts. He claims it allows for better understanding of linkages and patterns over time.

But what had my jaw drop was that he used Claude / ChatGPT to create the cards! Of course. Just throw in the material you are learning into Claude, ask it to create SR cards for the material and you’re done.

It may need some editing or creating additional cards, but it’s generally good enough. It can even output in Anki file format. Rope in an AI and the amount of knowledge you can include in your SR system just explodes.

While researching this piece, I revisited Alexey Guzey’s original post that introduced me to the concept. I didn’t remember this, but the first thing he talks about is to use SR for instilling novel thought patterns.

Front card: saying no

Back card: If I want to say no, I will stop and make sure this is not just status quo bias (coz it probably is)

Status quo bias sucks. Does this one solve it? Beautifully so. Although several months had passed before it finally kicked in, the number of times I noticed saying “no” out of status quo bias, then having this thought come up and make me retroactively reverse the initial “no” is in the tens already.

Will this work? I am not sure, but between how Dwarkesh is using SR and this, I feel the case for having my own SR system is growing stronger. Hopefully, I’ll have an update in a few weeks.

How do you train?

Tyler Cowen’s post inspired by David Perell planted another bug in my brain.

“Athletes train. Musicians train. Performers train. But knowledge workers don’t.”

Recently, one of my favorite questions to bug people with has been “What is it you do to train that is comparable to a pianist practicing scales?”

If you don’t know the answer to that one, maybe you are doing something wrong or not doing enough. Or maybe you are (optimally?) not very ambitious?

Here’s his answer. Of course, I didn’t know the answer to that one.

Watching Andy Matuschak study Quantum Mechanics with Dwarkesh Patel (who else?) led to a few realisations.

Having the right tools and workflows and being good at them is a huge productivity unlock. In the video, Andy uses custom software that allows him to do much more, faster. He reads, creates SR cards, and reviews material on the fly.

He also knows how to put mathematical symbols in his notes, making useful cards so much more quickly.

What is the difference in learning between someone manually revising from a textbook before an exam (as I used to do) and someone using a custom SR system, AI-generated SR cards, and a workflow to push those cards daily? Huge.

This year, I plan to practice getting better with the keyboard. How long can I work on my desktop without touching the mouse? To switch between windows and apps, and use functions within apps?

Sure, I can do Ctrl + C and Ctrl + V. But can I effortlessly do Alt + E + S + V + F (Paste Special > Formulas in Excel)? Can I do 50 other menu actions across multiple programs? Can I set up workflows for routine processes and use AI to automate?

The second thing that struck me was the depth of his learning approach. Sure, it’s a tough topic, but he spent 20+ min on the first page, asking questions, drawing connections, testing himself, and revising. I don’t recall engaging with my coursework so diligently.

The last thing that occurred to me is how important and hard it is to learn these hacks/tricks/workflows. Some people just know and others never do.

Here’s a great resource for this “tacit knowledge.” I’m glad someone recognised the importance of the concept, named it, and created a knowledge base.

The last two are a great if you are thinking about how to incorporate Generative AI in your work or life.

A hack I learned recently is anytime I have a question I am too embarrassed / worried / lazy to ask the right person or even google for, I just ask Claude or ChatGPT. Here are some examples of questions that would have gone unanswered but for the chatbots.

A piece I was reading mentioned that Starlink's low earth orbit satellites have a useful life of 5 years. I wondered what happens to them afterwards. Sure I could have googled that1 and probably found the answer somewhere in the top 3 links2. But having the bot answer your exact question has a lot less friction than doing that and I find I am asking a lot more of these questions.

I recently started reading Cloud Atlas by David Mitchell, apparently a modern classic. After the first 10 pages I was really struggling to understand what was going on and ordinarily I would have just dropped the book.

Instead, I asked Claude "I am reading the book Cloud Atlas. I am struggling to understand the theme through the first few pages." I got this response: "The book is structured in a unique way, with six nested stories that span different time periods and genres. Each story is interrupted halfway through, only to be concluded in reverse order in the second half of the book."

No risk of spoilers and I suddenly felt more comfortable navigating the book.

Claude has already changed my mind on multiple questions and I am sure will keep doing so. I hereby declare 2025 the year of asking more questions.

Did you know that (in iOS at least) this button allows you to dictate text? I used to think it was to send a voice message (don't do that) but no, this types it out and does a pretty good job of it too. This deserves to be used a lot more.