
"Stealing the Future": Effective Altruism is a financial fraud. Tech + Finance Analysis @ZKZeroKnowledge 'Hardcore' - NYMag X-Fortune, CoinDesk PhD @uiowa
Bushwick, Brooklyn
Joined April 2009
- Tweets1K
- Following6.8K
- Followers17.4K
- Likes234K
Pinned Tweet
We're finally waking up to Effective Altruism. I wrote the book on it.
STEALING THE FUTURE — SBF, elite fraud, and the cult of techno-utopia. Out now.
amazon.com/Stealing-Future-B…
And check out my weekly newsletter, Dark Markets: davidzmorris.substack.com/
David Z. Morris retweeted
BREAKING: Robert Edward Grant claims Claude may have plagiarised his work to help prove the Riemann Hypothesis.
Grant says he has written to Dario Amodei about parallels between Levent Alpöge’s proof and the approach he was developing independently. The full 16 page letter has been posted here on X.
Similar questions came up earlier this year over the Navier–Stokes problem and the counterexample to Gromov’s soficity conjecture. In both cases, several mathematicians claimed that their own chats may have been used to produce the proof.
It’s virtually impossible to tell whether a particular chat shaped a model’s answer. So we will probably never know whether these particular claims are true or false.
But what’s becoming clearer and clearer is that AI labs are training their models on our ideas at the frontier of human knowledge.
I think this is particularly problematic for at least two reasons:
The first is about intellectual property. If an idea I share in my chat turns up in your proof, where do we draw the line between what’s mine and what’s yours?
The second is even more worrying. Frontier AI labs sell our knowledge back to us via paid subscriptions.
They are literally privatising human collective knowledge.
David Z. Morris retweeted
Replying to @lionel_trolling
In fact, it's well established that correlation doesn't truly imply causation: it just opens a window for speculation and further research.
This is what spurs randomized trials in medicine.
Often, when causation looks flimsy, researchers tweak the data to make it seem stronger.
David Z. Morris retweeted
This from @lionel_trolling’s newsletter is an interesting, high stakes question. I suspect that a lot of people, in their gut, think everything is ultimately a computation problem open.substack.com/pub/johnga…
David Z. Morris retweeted
It is of course cool that the machine can do extremely advanced mathematics, what I object to is the *philosophical* position, baked into much of this discourse, that AI doing advanced mathematics is proof that a conscious Machine God has been created or is in the offing.
David Z. Morris retweeted
A lot of legitimately smart people who are STEM types have spent their whole lives believing that their domain of knowledge is the hardest and most specialist and they’re having a difficult time coming to grips with the fact that their work is the easiest stuff for Machine to do.
To counterbalance some of my negativity, I have been on the fence about some of Tyler's work over the years but he has really been killing it lately
I’m sorry but I refuse to be surprised that if you throw a bajillion dollars at Machine it turns out to be good at Number. Sure, that makes sense! So much of the hysteria stems from the fact that a lot of silly people also think Thought is Number and Brain is Number Machine.
2
108
David Z. Morris retweeted
We wanted to understand the significance of the proof of the Riemann hypothesis. So we talked to three Atlantic staffers with social science degrees at a Pain Quotidien on 55th and 8th.
David Z. Morris retweeted
Sam later gave Lily and her partner at Modulo Capital something like $300m, partly it seems in an effort to marginalize Caroline by replacing Alameda with Modulo as a primary market-maker on FTX.
David Z. Morris retweeted
Barbara Fried being a total narcissist who talks over her son is on the nose, as far as I have been able to figure out.
If anything, the chracterization of Barbara off the bat downplays how much of a psycho she has come off as over the years.
David Z. Morris retweeted
Katy Perry's presence was tied to Sam's dishing out huge amounts of money to Michael Kives, a bundler for Hillary Clinton.
David Z. Morris retweeted
Replying to @davidzmorris
"ASI", woof
Just... woof
She might as well be posting from the Grokbot pickleball tournament
David Z. Morris retweeted
They didn't verify the solution. Another ai did. Burden of proof. Until someone can succinctly explain the proof, I have no obligation to believe they have one.
"Predictions are falsifiable" isn't just an epistemologically seductive falsehood, it's a politically motivated claim to power.
If you're interested in delving into this, I'm starting a book club to read Carissa Veliz' incredible recent book Prophecy over at Dark Markets.
Replying to @lionel_trolling @jonathanchait
Dude you're smarter than this, the whole reason to make predictions is to test whether your models of the world are falsifiable. We've been making testable predictions for years and gotten a lot of shit right and beaten markets/pundits/other models/the CW etc.
David Z. Morris retweeted
imo it’s worth also taking seriously the blatant gap in capacities that is ai not being able to satisfyingly explain the proofs it produces, it isnt obvious to me that we should expect that to be the case
David Z. Morris retweeted
That's an absolutely absurd and indefensible comparison. The climate is a natural system that is clearly quantifiable according to physical laws. Politics is a highly contingent, context specific and embedded in human institutions.
Replying to @NateSilver538
I've compared the tendency to climate science denialism, because it involves an alliance of extreme claims with nitpicking allied experts theatlantic.com/ideas/2026/1…
The idiots arguing over the settlement of these positions is going to be glorious to behold
BREAKING: Odds of the AI bubble bursting this year surge to 69% as the world's 10 largest companies lose over $2,900,000,000,000.00 in a week.
🔗 polymarket.com/r/pDNVcvT
David Z. Morris retweeted
Replying to @jonathanchait @NateSilver538
There's no such thing as "election science." There is a huge difference in rigor between the natural and social sciences
David Z. Morris retweeted
They aren't falsifiable. If you make a good prediction you say, "Yep our model is right," if not, you say, "Well we did say it had 20 percent or so chance." That's the definition of being unfalsifiable. What would falsify the entire model?
Replying to @lionel_trolling @jonathanchait
Dude you're smarter than this, the whole reason to make predictions is to test whether your models of the world are falsifiable. We've been making testable predictions for years and gotten a lot of shit right and beaten markets/pundits/other models/the CW etc.




















