Balaji

Yann LeCun

At this point, FB really should consider open sourcing it officially. It would still earn goodwill, just like they did with React. @ylecun https://t.co/Xfveh8ihsX

Meta’s LLaMa weights leaked on torrent and someone put up a PR to replace the google form in the repo with it 😂 https://t.co/Hu2EUH6jmm

ceo & head of engineering @replit—build faster on https://t.co/oCT5NH3qPT

Amjad Masad ⠕

Facebook did something really good here — they developed a cutting edge model and released it to the world.

If they just take away the Google Form and lean into the open version, they'd earn rightful praise.
https://t.co/5sDr7xbEkJ

Immutable money, infinite frontier, eternal life. #Bitcoin

We're starting to see physical social networks.

A physical social network is a cloud community that periodically materializes on the land. That means people who met for the first time online start meeting routinely offline. https://t.co/iPvVFtT0pX

4% of the protocol in one place IRL! https://t.co/X8MEMpcb5x

Dan Romero

I made the table below in 2013, as an accompaniment to this article: https://t.co/XKxhg8q6v3

I think it holds up well. We have seen cloud gatherings of steadily increasing scale and duration. And there is no physical upper bound. https://t.co/kmZpwVyGGS

Remember the “chip ban”?
China is apparently getting ASML blueprints anyway.
They are spying their way to technological independence.
https://t.co/2wI66r5w9s

We’ll see whether China gets there.
But the espionage was predictable… https://t.co/yRlbsWTs4K

1) This thing does what we imagine
2) So let’s imagine all the ways it could kill us

This is of course partially a joke, because LLMs don’t *actually* do everything you can imagine.

But the serious part is that sometimes obsessing over x-risk causes it. https://t.co/UVUgx5E0vg

Good thread. I agree with much of this.

Though I do think there are many areas of the economy where AI-aided is way better than human-only.

We’re already seeing improvements in search, coding, writing, art, and design. https://t.co/NX1DLxE88a

My Twitter timeline is full of panicked takes about imminent AI apocalypse and certain doom.

I think this is starting to get overplayed, and so I want to make a long thread about why I'm personally not worried yet.

Get ready for a big one... 1/n

Recovering economist turned biotech VC. Lossy compression not recommended. No, we have not picked all the low-hanging fruit yet. My other half: @diviacaroline

William Eden

ABCDs of apocalypse

AGI: AI may kill us, so halt progress
Bitcoin: hyperinflation may ruin us, so end all states
Climate: warming may harm us, so halt growth
Democracy: authoritarians may control us, so risk nuclear war

The risks are uncertain.
The “cures” are certain. And bad. https://t.co/974iuhYpEC

There are so many technically possible secular apocalypses: Carrington Events, nuclear exchanges, designer viruses, even quantum computing.

But those working on preventing them sometimes help bring them about. The COVID lab leak may be an example. https://t.co/UVUgx5EykO

One observation: the AGI instant death scenario is a secular match to Pascal’s Wager.

Yes, if an omnipotent being existed, it’d be dumb to not fear it and devote your life to it.

Even if you think the probability is low, the downside is infinite right? https://t.co/974iuhYpEC

The counterargument is: we can see developments in AI accumulating in a way we don’t for God.

The counter counter: yes, but many other apocalypses have accumulating evidence too. And many religious people believe all is evidence of God’s will, eg physics is evidence of designer.

I may write more on this.

But in short: the hard takeoff + instantly invent nanotech + kill everyone scenario has many unproven steps.

Like — nanotech doesn’t exist.

And — superintelligence isn’t omnipotence. Adversarial AI exists.

And — embodiment isn’t trivial.

In many ways, the AGI debate presumes the consequent.

If you *assume* an omnipotent being, of course you can’t outsmart it. Then you let it take every sci-fi tech with some calculations around it and instantly push it to production.

Why doesn’t AGI instantly invent time travel? https://t.co/7eAHD0ceWL

Another issue is equating “big” with “infinite”.

It’s superintelligent…thus omnipotent.
It’ll get smart fast…infinitely fast.
It outsmarts one human at one thing…so it’ll outsmart all humans at everything.

Omnipotence can’t be beat.
Finite systems often can, though. https://t.co/Iz5LWPUhij

Yet another is equating the gray goo scenario with the “persuade some humans” scenario.

In the scenario where it controls robots to invent as-yet-uninvented nanotech the AGI can act alone.

Once you have a few cultists listening to AI whispers, it’s more like Aum Shinrikyo. https://t.co/NuUBHhiWP4

There are many examples through history where people listening to whispers from their god executed terrorist attacks.

The AI-persuades-humans scenario looks like secular Al Qaeda.

The problem then isn’t omnipotent AI — it’s just an LLM — but an impressionable, fanatical human. https://t.co/7WfudE5QXY

Another failure mode for AGI-instantly-wipes-us-out is like the Drake equation.

You can get wildly different results from reasonable-seeming assumptions. But the observed answer so far is zero aliens.

Point: could a seemingly plausible scenario actually be zero probability? https://t.co/kWRqQCImKT

With all that said, consider this 3 point scale.

1) AI is a stochastic parrot
2) AI will create billions in value
3) AI may kill us all

I’d be a 2.5, but not a 3.

I don’t think hard takeoff + gray goo is likely for reasons outlined above.

But I do think robot war may happen. https://t.co/mX8gUaS8CO

Political backlash to AI is underrated.

Will you *actually* be able to make billions disrupting the Democrat base from within the bluest city in the bluest state in the union, when SF and CA are both already facing enormous deficits…or will they tax and regulate it to death? https://t.co/8e5jNNRo93

Just for perspective, California is reporting a $20B+ budget deficit and says that might even triple.

And 500k people have left California for places like Texas.

So where will they find $20B?

Ah. Those evil AI guys in SF with billions, taking all the jobs. The obvious target. https://t.co/44Q3Ghya2X https://t.co/XFhsZP3sEN

California tried to build high-speed rail.
The cost rose to $100B.
The timeline rose to ~100 years.
So, there is no high-speed rail. https://t.co/e8VvyMJzAr https://t.co/935rVooX7m

The U.S. can't build housing. Or transit. Or solar power. Or transmission lines. Or semiconductor fabs. 

We are the Build-Nothing Country. 

And this needs to change, and it needs to change fast.

https://t.co/VcvgKMiAZP

Writes about economics, tweets about rabbits.

Subscribe to my Substack: https://t.co/N0VDjJuCZ7

Email: noahpiniontwitter@gmail.com

Noah Smith 🐇🇺🇦

The US establishment has inherited a country they could never have built. The founders left the building long ago. 
https://t.co/mnPr3zkz7g https://t.co/aMqrrBwOFq

A society that can’t build from scratch will eventually find it hard to maintain.

Because if you can’t build every part of a system, how are you going to build the spare parts for that system?

Something will eventually break that you can’t fix. And then, idiocratic decline. https://t.co/lNKAQNEgiW

The future is already here — it’s just oppressively regulated.

The ultimate example is biomedicine, where sequencing costs plummeted even as drug costs soared.

[1]: https://t.co/wU6k5Yo6oj
[2]: https://t.co/FA8FPjnSK2 https://t.co/TAIuQj7vJX

It’s true that this trend has shown some signs of flattening out. But a full reversal is still orders of magnitude away. https://t.co/YY4EQwHBq3 https://t.co/kXSKE1rgFw

Why the US keeps backing Pakistan is anyone’s guess. Bin Laden was found less than a mile away from the Pakistan Military Academy. https://t.co/75avkyKUCd https://t.co/DTwQOnswA0

"will support...national security objectives of the US by allowing Pakistan to 'retain interoperability with US and partner forces' in ongoing counterterrorism efforts."

How many times has Pakistan used an F-16 in a counterterrorism operation? This is not a rhetorical question. https://t.co/552oV8trPL

Research Fellow, South Asia at The Heritage Foundation. Author, Cold Peace: China-India Rivalry. Author/Editor, Asia's Quest for Balance. RT≠endorsement.

Jeff M. Smith

The bin Laden / PMA map is reminiscent of the Wuhan / wet market maps.

Let’s call them…suggestive. https://t.co/R6SGYijoph

Balaji Srinivasan

Is financial privacy legal?
Should HTTPS for cryptocurrency be applauded?
And can we actually build an alternative to surveillance capitalism and surveillance states alike?

That's what this is about. https://t.co/71Z6bFPm4o

1/ Coinbase is funding a lawsuit brought by six individuals challenging the Treasury's sanction of the Tornado Cash smart contracts. https://t.co/DkMHtyt0zR

Co-founder & CEO at @Coinbase. Creating more economic freedom in the world.

Brian Armstrong

Remember, the US government once tried to ban encryption. Then they made it mandatory.

[1]: https://t.co/SN1oqwAA6L
[2]: https://t.co/BTQ7pj9KmB https://t.co/JCLzWz6KZC

The digitization of everything changed the defaults. It's too easy to casually invade someone's privacy. The encryption of everything will change the default back.

Make surveillance hard again.

The good news is that the establishment has much diminished soft power. They do not convince anyone who isn't already convinced.

The bad news is that the establishment retains much hard power. They do not convince, so they will coerce.

How do we know soft power has plummeted?

The stock market is down, and that's been much discussed. But media engagement is also way down, and that's been less discussed.
https://t.co/pwD0PwgVbE
https://t.co/plSImC2TQo https://t.co/kmCbYN0tae

The last (unannounced) pivot was from the war on terror to wokeness. The next pivot is already underway.

It's the pivot from wokeness to statism.

Just raw force in the name of the flag, against global dissidents and decentralists. Causing realignment.
https://t.co/JxkjgcuEPG

@lessin The subgraph pictured is just an example, one could do a more rigorous version.

The human judgment comes in the selection of the initial “training set”, namely the nodes one labels as part of the establishment.

(Note that there are also multiple counter-establishment groups.)

@balajis's Latest X Threads

Read threads by Balaji and get notified about future ones.