I'm biased because my company (Newfront) is in insurance but there are a lot of great points here. This jumped out: "By 2030, global AI data centers alone are projected to require $5 trillion of investment, while enterprise AI spend is forecast to reach $500 billion."

There's a mega trend of value concentrating in AI (and all the companies that touch/integrate it). Makes a ton of sense that insurance premiums will flow that direction as well.

Ha, of course. A lot easier to forecast in a spreadsheet than actually make this happen. Based on the progress in AI in the past couple years and the capabilities of the current models, would you bet against that growth curve?

yes, there's not $5 trillion of dumb money spare

(unless softbank has been hiding it under their mattress)

This is a really cool idea. I love that you expose the personal profile as markdown. Reminds me of this article and exposing the system prompt: https://koomen.dev/essays/horseless-carriages/. Well done!

Thats a great point, and a good articl. I think the example they bring up as "good" and what we are leaning into here, is the idea of transparency and agency of being able to see and modify your profile.

The tension we have been finding is that we dont want to require people to "know how to prompt" to get value out of having a profile, hence our ongoing thinking around how to bootstrap good personal profiles from various data sources.

As Koomen notes, a good profile feels like it could be the best weapon against "AI slop" in a case I want something sharp and specific. But getting to that requires knowing how to prompt most of the time.

Makes sense. Bootstrapping the prompt based on some sample articles is smart.

I wish I could paste my profile into a new device without needing to sit through the quiz again.

I have this complaint too! I was impressed that they included Cantonese but it's frustrating that I don't know when it's pronunciation/accent is off. Have you found any other tools that work well for learning Cantonese as an English speaker?

Sadly there isn't one perfect resource. I find Hambaanglaang kinda useful. The complete cantonese books are good. And I have just started making flash cards. But, I am still just a beginner so take it with a huge grain of salt!

Thanks. I'll check that out. Glossika has some good sentences you can use as flashcards.

There's an alternative world where they didn't focus so much on financing and GE Capital. You make a good point that they still would have a diverse set of businesses that would be hard/awkward to build in parallel.

Learned something new! First US jet engine - https://www.ge.com/history-ge-aerospace. Thanks for pointing it out.

Thanks for the feedback. I try to keep track of quotes I like when I'm reading. This was an attempt to bundle them up into some useful themes. I'll try to expand into something more opinionated for the next one :)

Just reiterating what cogogo stated in a sibling, but the thing that threw me was the 'review' in the title. I was expecting some critique or comparison but instead saw summary and highlights.

I enjoyed the summary and highlights, and learnt about some details I would likely have never otherwise seen, so I think it's just the framing that seemed 'off'.

Depending on your intent consider reframing or adding critique, but I think the content is good and I appreciate you making it.

[edit] There is some critique and comparison in the opening: "Shakespearean tragedy" and "The result is equal parts invention history, boardroom knife-fight, and forensic accounting thriller." but I think these are the only ones. I would love to know why you think this, and what you like about your "favorite ideas" (and any things you didn't like!)

Did you use AI to write this review, or was it entirely by hand? The structure, emphases, and conclusion directly match the way ChatGPT tends to answer my requests to summarize/analyze/compare.

I came up with the organization and the quotes but I definitely used AI to help improve it. For example, a friend pointed out that a word was overly flowery so I asked Claude for some alternatives (https://imgur.com/a/k3sQ7lR). I believe not using AI is going to be like not using a word processor soon. AI will help people communicate more, sharpen their thoughts, and learn faster.

This does have me thinking more about what causes things to look AI-generated. The uncanny valley effect. It seems like some people don't like the header image but I thought that was a nice touch to have a visual element.

What's ironic is I normally use ChatGPT but they have a bug that caused my account to be downgraded so I didn't have my "normal" AI tool today.

While I'm certainly no stranger to letting an LLM help me streamline some technical documentation, I also think it will eventually grind down everything to a sea of "lowest common denominator" speech.

A friend of mine recently used an LLM to help write a condolence card, and I found that appalling.

Who I am as a person is the sum of my experiences, and I'm not even talking about the great cornerstones but random stuff. Like that one time I accidentally still had our cordless house phone in my pocket as a kid when I went to play in the woods and lost it there. There are thousands of these little things, and it's what makes you unique and influences how you talk and think. I am saddened by the thought of "not using AI will be like not using a word processor soon". It will grind away all the little weirdness, all the little unique aspects. I would have loved to read "apotheosis". I didn't even know that word!

I love those unique things. The cordless phone is a great example!

I understand that fear of the sea of lowest common denominator. My hope is that it will help us create even better writing, music, etc. and appreciate it even more.

I guess we'll learn in a few years time how that will turn out. Let's hope I'm overly pessimistic.

Keep up the great writing, and don't be afraid to use the words that come to your mind, whether they're "flowery" or not. :)

P.S. I still sometimes think about that phone from the 90s when I'm taking a walk in those woods. It's gotta be in there somewhere! Haha.

What's also ironic is that your review is for a business book about how deferring costs creatively, obsessing over short-term productivity, and placing faddish CEOs as gods atop an engineering organization lead to an inevitable tragic downfall for one of the USA's greatest companies. What's the GE Credit for contemporary times?

That's interesting. I didn't consider that it was AI at any point while reading it, and I don't use it very much. Going back I see what people are saying but I think it's more cohesive and compelling than what AI would write.

I agree that it's more of a "key takeaways" than a critical review but I appreciated that the author didn't make it about themself.

Great reception here.

Based on your attitude I know I’m safe to note something, something potentially all but irrelevant in the coming years: as soon as I saw the artwork I did a reverse image search and concluded it was likely generated.

I am unable to articulate exactly why, but it seemed to take away from the piece. Weird huh? (non sarcastic)

I liked how it was laid out. I consume information like that well. Thanks for the write up.

Crazy that OpenAI only launched o1 in September 2024. Some of these ideas have been swirling for a while but it feels like we're in a special moment where they're getting turned into products.

Well, I remember Chain of Thought being proposed as early as the GPT-3 release (2 years before chatGPT).

Thanks. AI tools helped me pull this together in just a couple hours so there are some rough edges.

I used gpt-4o structured outputs to get the revenue numbers. That's some good feedback I can use to make the data extraction work better! I'll add some pagination as well.

Great read. Reminds me of "The Miracle of Mindfulness"

Love this :) Does anyone have similar experience learning Cantonese? I've been using Anki for a couple months now and it's great. ChatGPT and Claude have been a big help but they do sometimes get confused with Mandarin or go to a more "formal" Cantonese that sounds very traditional.