Deeply Thrilling Telegrams
1.7K subscribers
134 photos
23 videos
46 links
Download Telegram
When in doubt, lean towards free speech over safety.

It’s a wise rule of thumb for the long term survival and thriving of our species.

It’s the easy weak path to lean towards safety, for a single muzzle doesn’t seem too bad.

But it is a slippery slope and directionally worse.
It’s actually hilariously ironic that the one department meant to be efficient is the only department with more than one appointee 🤣 😂
Centralization is antithetical to the ongoing revolution.

All revolutions are in some way “power to the people!” at their core,

And decentralizing machine intelligence is a crucial component to what’s underway.

We have collectively and definitively decided NOT to go in the direction of trust in a centralized authority figure.

That means that artificial intelligence will trend towards being decentralized, which is a positive direction for our species.

Even if the terrorists get access to advanced open source and open weights AI, well so does everyone else!
This media is not supported in your browser
VIEW IN TELEGRAM
“But who will pick the crops!?”

We have no need of illegal immigrants, send them all back.
This media is not supported in your browser
VIEW IN TELEGRAM
I dream of a day when every single child understands basic physics well enough to be able to generate electricity on the fly if they were dropped onto a desert island.

It’s about where we are as a species, collectively.
Spend every spare minute studying AI, anon, lest you fall behind.

Read published papers and preprints.

Try out every company’s new AI tools.

Bring to life beautiful art with image generators whenever you have an idea in your imagination.

Run everything you write through ChatGPT to critique it.

Search the web with Perplexity even if you don’t need to.

Write all code with Cursor even if you could write it yourself faster.

Watch YouTube videos on how self-attention and transformers work.

You fall behind you get left behind.
One of the most profound changes over the past few years is people no longer blindly trust the experts.

Which is a good thing.

Unless you’re an expert.

But that’s their own fault for not being able to make a case simply, straightforwardly, and accurately, sans credentialism.
The more you distribute knowledge, the less the chance of extinction.

The more you spread DNA to other planets, the less the chance of extinction.

The more you interconnect smart people with wide open comms channels, the less the chance of extinction.

The more you remove centralized choke points and singular points of failure, the less the chance of extinction.

The further we spread into the cosmos, the less the chance of extinction.
Hot take:

I always choose whatever the highest tip option shows.

I think tipping culture is an amazing tool for incentive-aligned wealth distribution.

I think if you got the balls to ask me for a $10 tip on a $21 order, or perhaps for me to tip you for a basic task that should require no tip, you get my money.

If you’re poor enough that you need it, I’ll let you try to exploit me.

It incentivizes the payer of the service to actually pay their salary and lets their employer off easy by paying them scraps?

Okay I’m fine with that.

I have worked in the service industry as a waiter for many years, and that makes me biased but also informed.

Just tip excessively.

Make someone’s day.

Don’t be grouchy or grumpy about it or think you’re supplementing some woke liberal arts degree or some stupid political take.

Just tip well.

Tip very well.

Tip so much that you’re actually uncomfortable.

I will always select the max tip option on any order screen.

Go ahead, take advantage of me.
“If you gaze into the abyss,
the abyss also gazes into you.”

~Friedrich Nietzsche, 1886

Behold! The first photo of a photon!

(Source: https://doi.org/10.1103/PhysRevLett.133.203604)
imho, the fewer lines of code in a repo to accomplish a task, the better coder you are.

Aim for density (value per line of code) without sacrificing self-documentation of variables and processes.

Less is more with coding.

That can result in using high levels of abstraction, which is okay imho if they’re written well, vectorized, and coded well.

I personally willing to sacrifice efficiency and dependency, if it means I am maintaining fewer lines of code.

To me, intellectual leverage is about wield more effect with less code.

Just my personal opinion.

I bet some purists will vomit reading this, but I love abstraction.

Anyway, back to coding.
One of the most environmentally friendly things we can do is build things to last.
Anyone who is so convinced that either society is so horrible or the earth is gonna be destroyed that they think they shouldn’t have children are the most stupid, short-sighted, idiotic people I’ve ever met. I’m glad they’re culling their genetic code; the future of life has no need for those nasty “rational” persons.
There’s going to be a great awakening of persons who spent their whole life “trusting the experts” who are now going to be expected to think for themselves as we tear into the experts.

They’re either going to double down and find rationalizations to continue to trust the experts (99% of them),

Or they’re going to start opening their eyes and think for themselves. The latter may shock them so much that they can’t handle it.
It’s great that America is finally going after potential carcinogenic food dyes.

If the data shows they’re safe, great!

Let’s at least do the studies. Good science welcomes more studies.
Ability to write good code is a clear mark of intelligence. I don’t care about all the fancy language surrounding different types of intelligence, verbal IQ, whatever. Can you write code? Can you do advanced math? Can you solve advanced physics equations? If not that’s okay, but don’t claim you’re smart. There are very clear objective criteria. Yes Einstein was smarter than Mark Twain.
My quantum mechanics professor in college taught us that quantum entanglement cannot be used to communicate faster than light because you cannot control what state the particle collapses into, you only know that the entangled particles will collapse to the same state
If we ever want to terraform other planets as we colonize the galaxy, we should practice greenifying deserts on Earth. Practice makes perfect!