🧠 🤯
karpathy speaks like someone who’s running a mental compiler in real time with minimal interpretive latency & almost zero runtime garbage. he’s not verbose. he just threads complexity into compressed lossless statements.
most smart people can be dense, but they lose clarity. karpathy keeps clarity while cranking the bitrate like an llm tuned with perfect temperature control.
it’s so beautiful to listen to that i had to hear this three times already. i might add a fourth.
ML retweeted
The biggest AI Agent opportunities are going to be in categories of work where there could be a 10X or 100X increase in demand if the cost of executing certain tasks went down dramatically.
The key is to find categories of work where there’s a significant amount of non-consumption simply because doing the work before was too expensive or time consuming.
It turns out that these spaces are basically in every single market. The easiest way to figure out these markets is to find out which verticals have things that customers have always wanted to do but never get around to.
If you talk to lawyers, they’ll complain about how time consuming reviewing every contract is, and how many contracts are sitting around that they don’t know anything about. If you talk to engineers, everyone knows how endless the backlog of mindnumbing work there is to get done that isn’t core feature development. If you talk to marketers, they always would love to launch more targeted campaigns to more markets.
In every one of these areas, AI Agents will bring down the cost of doing this work dramatically, which means the work can finally get done.
And importantly, all of these tasks are tied to broader workflows, which is desirable given the need for ultimate human review of most output. There are endless opportunities like these to go after right now.
ML retweeted
“The highest form which civilization can reach is a seamless web of deserved trust. Not much procedure, just totally reliable people correctly trusting one another.” — Charlie Munger
YC is this. Combine that with decades of compounding and we as a community will go far together
ML retweeted
On January 1, 2014, I wrote a letter to my future self just before graduating.
Today, 10 years later, I opened it.
This hit me hard.
Everyone should do this.
ML retweeted
I'm not an effective altruist, but I find the recent genre of pieces like the below a bit strange. Essentially all of the major AI lab leaders agree that AI potentially poses enormous risks[1], as does a majority of the US public[2]. It's not a crank view. EA was simply among the very earliest organized groups to perceive and act on this risk. In addition, EA's contrarian concerns have elsewhere been shown to be reasonable: long before COVID, EAs stood out for their worry about major pandemics. Assessments that deride EA as a cult while failing to acknowledge these counterintuitive successes strike me as unreasonably uncharitable. If asked "which 2018 community now looks most prescient based on how the intervening 5 years unfolded?", I think it's hard to come up with better nominations than EA.
politico.com/news/2023/12/30…
[1] nytimes.com/2023/05/30/techn…
[2] rethinkpriorities.org/public…: a majority rate the proposition that AI will cause human extinction within 50 years as at least "slightly likely".
ML retweeted
5/ Buffett said: “If you want your kids to have certain values, it's important that you live those values and talk about it.” This applies not just to parenting but being a boss. If I don’t practice what I preach, my employees will not take me seriously.
ML retweeted
4/ Buffett said, "If you want to know how to live your life, write your obituary and then reverse engineer it so that you can live up to it.
TIL "vegan whey" is made by altering the DNA of bacteria to trick them into making whey protein for you! You get the benefit of milk without the inflammation of lactose
perfectday.com/blog/how-we-t…
ML retweeted
In a (future) world in which everyone has access to personal AI assistants, human knowledge & intelligence become less important than motivation, moral compass & ability to listen.
Just like political or business leaders who are surrounded by advisors smarter than themselves.
ML retweeted
Elon is going to be voted out of the poll he said he would abide by.
He then said no one wants the job who can do it…
Then seemed to be open to @lexfridman running it. Lex has a popular podcast but only prior professional job was 1 year at google.
“A recession in the next 12-18 months appears to be a foregone conclusion… coincide with deterioration of corporate earnings and investor psychology”
- Howard Marks
oaktreecapital.com/insights/…
Evidence banks are retreating from risky financings
Banks are nowhere to be found in recent LBO financings
Thoma Bravo's $8 billion $COUP deal includes a $2.6 billion loan led by Sixth Street Partners and including HPS Investment Partners, Oaktree Capital, Apollo Global Management, Blackstone and more
bnnbloomberg.ca/sixth-street…
ML retweeted
Just how bad is the crypto collapse?
My landlord friend in Dallas just asked me if I know anything about bitcoin miners:
“This bitcoin mining company defaulted, moved out, and left all their equipment behind.”
200 Antminers 🏃🏽♂️💨
ML retweeted
"Whichever the outcome, the world wins. If Musk manages to fix Twitter, we're left with a better Twitter. Great! If Musk manages to burn down Twitter, we're left with a world free of Twitter. Great!" world.hey.com/dhh/heads-we-w…
ML retweeted
"American companies will never be able to resist the demands of their intelligence services. It doesn't matter if their servers are in Virginia, Paris, or on the damn moon. Europe should either come to terms with that reality or raise a real privacy wall." world.hey.com/dhh/american-d…
ML retweeted
The only effective way to update processes in a company:
1. Add it to a place that ALL future employees see.
2. Announce it on Slack/email.
The biggest mistake is doing only 2.
We just coined "permanent publishing" at @unit_co_ to reinforce this. Would you use a different term?
