Cosmic anger, the problem with muting your mic, and the funniest math joke...
TNW View in browser →

Punching a hole in the universe and sticking your middle finger through it

Howdy folks,

You ever get so mad that everything tastes like rage and aluminum for a week? No? Just me? 

I was thinking about energy this morning. It takes a lot of energy to sustain anger. It's exhausting. 

And, frankly, I don't think it's very good for the universe. 

Last summer I wrote about how there could be a doppelganger universe full of white holes on the other side of ours. It's a pretty wacky concept. 

But there's no scientific consensus ruling the idea out.

I was thinking about all of that when it hit me: every time I'm sitting there mad enough to chew iron bars and spit out railroad ties, it's technically within the realm of possibility that there's an opposite-universe version of me out there having a great day. At my expense!

It makes me want to punch a hole in the universe and stick my middle finger through it. 

But then I thought about the last time I was happy.

And that made me smile.

And I thought about what that meant for the doppelganger version of me.

And my smile got bigger.

gring smile

Perhaps the best revenge truly is living well. 

Tales from the arXiv archives

Chain of Thought Prompting Elicits Reasoning in Large Language Models

I get excited about pre-prints. I can't wait to find out what happens next. And there's no better way to do that than to scroll through the arXiv archives

But this paper has me even more excited than usual. 

Remember a few weeks back in the "It's time for robots to start paying taxes" edition where we talked about the Google Brain paper on, essentially, teaching common sense to AI? 

Well, it has a companion paper from the same team. In one part of their work they taught an AI to use common sense. In the other, they taught it to reason its way through problems using that common sense. 

Today we're talking about the latter paper. 

Per the team:

"This paper explores the ability of language models to generate a coherent chain of thought — a series of short sentences that mimic the reasoning process a person might have when responding to a question."

Combined, the papers present one of the most interesting views on brute-forcing coherence out of deep learning that we've ever seen.

Check them out here and here

What we’re writing


OpenAI’s new image generator sparks both excitement and fear I can't help but wonder what the non-curated outputs look like...


Muting your mic doesn’t stop big tech from recording your audio If you're not double-muted, you're not muted


Zuckerberg scrambles to monetize the metaverse amid fears over Meta’s future And just when you didn't think Facebook couldn't get worse...


Google says it classifies AI-generated content as ‘spam’ Take that, machines who want to steal my job


AI sucks at telling jokes — but it’s great at analyzing them Wocka wocka


Your brain shrinks as you age — and now scientists can prove it No joke here, these new brain charts are incredible

What we’re reading


AI predicts if and when someone will experience cardiac arrest (John Hopkins)


Quantum teleportation: The express lane for quantum data traffic (Phys.Org)

What's next for AlphaFold and the AI protein-folding revolution? (Nature)

Something profound from the internet

The more things change...

Our favorite video of the week

Wanna see what it looks like when an AI system pretends to think? Click to watch on Twitter:


Well, bye

Here's the funniest math joke I know:

Q: What does the 'B' in Benoit B Mandelbrot stand for? 

A: Benoit B Mandelbrot

See ya next week!

Any good?

How was today’s newsletter? Amazing? Awful?! Help us make it better by sharing your brutally honest emoji feedback 👇

Positive Meh Negative


Email with any suggestions, complaints, or compliments. If you don't want to talk to him, email his boss:

Facebook Twitter LinkedIn Instagram

From Amsterdam with <3