Pinned Tweet
many don’t have an opinion on something until they’re asked about it, at which point they cobble together a viewpoint from whim and half-remembered hearsay, before deciding that this 2-minute-old makeshift opinion will be their new hill to die on.
Robert Pattinson as James Bond makes sense. Daniel
Craig was a good Bond, but in his last 3 movies, Bond is beaten, humourless, and bitter. Time for a change.
youtube.com/shorts/4XTfAl7Cf…
14
This is Fine retweeted
The megatrends I’m spending the most time on:
1: AI + biotech economics
AI-designed drugs are reaching human trials. The bigger opportunity is whether better discovery translates into fewer expensive failures and more approved medicines per dollar of R&D.
2: The AI infrastructure buildout
Growing AI usage requires more computing capacity, networking, cooling and power. I’m interested in the companies solving the physical constraints on that expansion.
3: Domestic energy production, especially nuclear
Energy security is bringing renewed attention to domestic nuclear capacity. Reactor restarts, upgrades and fuel supply chains are worth watching alongside the development of new reactors.
4: Tokenization and onchain finance
Tokenized securities are moving into real transactions, including recent trades using DTC-tokenized assets. The opportunity is making assets easier to transfer, settle and use as collateral, with stablecoins developing alongside them as payment infrastructure. DTCC · BIS
5: Grid modernization and energy storage
Generating electricity is only part of the challenge; delivering it when and where it’s needed requires investment. Transmission, transformers, batteries and grid software all deserve attention as electricity demand grows.
6: AI agents and the economics of services
Businesses are starting to delegate complete tasks to AI. If that becomes reliable across more workflows, small teams could deliver services at a scale that previously required much larger organizations. Anthropic
7: Robotics and physical automation
Factories now operate roughly five million industrial robots worldwide. Better machine intelligence could make a wider range of physical tasks economical to automate.
8: Defense modernization
NATO allies have committed more than $40 billion to counter-drone capabilities over five years. I’m watching the growing importance of drones, sensors and the systems needed to detect and defeat them. NATO
8: Critical minerals and secure supply chains
Electrification increases demand for copper and other strategic minerals, while concentrated production creates supply risks. Mining, refining and recycling are becoming central to industrial security.
9: Cybersecurity and digital identity
AI is helping attackers operate faster and at greater scale while introducing new systems that need protection. Verifying identities, controlling access and securing AI agents should become increasingly valuable. Microsoft
10: Collectibles and Unique assets
In a world where production cost trends to zero, the wealthy will allocate to items with provenance and irreproducibility (classic cards, gemstones, gold, etc)
11: GLP-1s and metabolic medicine
These medicines are expanding beyond weight management into indications including cardiovascular risk reduction and liver disease. I’m watching how broader clinical uses reshape demand, manufacturing capacity and access to treatment.
12: Water infrastructure and reuse
Growing water insecurity is strengthening the case for treatment, recycling and industrial reuse. Reliable water supply could become an increasingly important constraint on where cities and industries can grow.
Plan it to make videos on each one
Here are 10 jobs that don't yet exist but could employ millions by 2040:
1. AI companion therapists
2. Brain implant calibrators
3. Climate engineering technicians
4. Synthetic organ fabricators
5. Space habitat maintainers
6. Longevity treatment specialists
7. Robot fleet coordinators
8. Neural privacy guardians
9. Gene editing counsellors
10. Quantum network operators
10
This is Fine retweeted
Replying to @fiveoutofnine
This is Fine retweeted
the cryptocurrency nerds are rambling today about new psychotic best practices they expect every user to follow in the age of AI
this pisses me off to no end. i don’t understand why this industry chooses panic over empathy every single time
the threat is VERY REAL. but the solution is not to teach end users to become cryptography experts whenever a new threat appears (which will happen at least once per quarter from here on out)
the solution is to encourage end users to outsource asset security to trustworthy, proven third parties
ideally just use the ETFs. if not, use trustworthy custodians like Coinbase. if you ABSOLUTELY want to be a self-custody caveman while facing the threat of super intelligent adversaries, AT LEAST pay a 3rd party to implement best practices for you (like bitgo, casa, privy/fomo if you’re an “onchain” degen)
i tweeted about this months ago. enough with the technicalities that nobody understand. crypto users are no longer limited to 300 nerds on a mailing list. get people to safety and stop behaving like cavemen.
the idea of your bitcoin or crypto resting easy in some secret location while you enjoy life not worrying about it is currently unrealistic
it was somewhat realistic before. but the cost of cybersecurity attacks went down exponentially, and continues to go down exponentially on a monthly basis as new, more capable AI models are publicly released
what this means is that your cybersecurity must now be actively managed. it cannot be passive. someone must worry about your shit every day and every night, actively dismiss old assumptions and attempt new approaches on a weekly basis to adapt to a dynamic landscape of new threats
so your options are:
1. be the one who is worried. assess developing threats yourself on a daily/weekly basis, stay up to date with new models and tools as soon as they’re launched, analyze your own security practices ruthlessly day and night. for most people, including most experts, i think this is highly impractical. but you’re welcome to try
2. store your coins with a competent third party service that you trust to do the worrying for you. they should have the resources, skills and incentives to actively manage the cybersecurity risk for you. if you don’t want to worry yourself you need to pay someone else to be worried. it is what it is
i’m not trying to shit on self-custody. technically you can have someone professional actively worry about your self-custody setup, that’s fine too
one day, the threat level will plateau and will stop increasing exponentially on a monthly basis, and then passive security will gradually become acceptable. but today is not that day
This is Fine retweeted
hey @grok, explain Vitalik’s post in plain language for a crypto holder who just wants to know if their money is at risk and what (if anything) to do.
I don't recommend anyone scramble to move their funds to new wallets today. But we should take the risks to cryptography from AI-accelerated math seriously, and minimize our exposure to not just quantum-vulnerable cryptography, but also potentially AI-vulnerable cryptography.
The core new area of risk from this viewpoint is, unfortunately, ML-DSA / FHE / lattices.
(and it's also another reason, along with quantum, why ECDSA might fall even faster than expected, hence the "fresh address" recommendation)
So far most people have been in the mode of thinking "elliptic curves broken, hashes safe, lattices safe". But there is a good chance that the concrete security of lattices will take serious hits from the next two years of AI math.
The basic threat model is: factoring is something that naively takes 2^(n/2) time, but over decades smart people have found and optimized number field sieves, and degraded that to 2^O(n^(1/3)), which is why RSA keys and signatures need to be ~400 bytes (and not 64 bytes). What if there are skeletons in the closet like that, both for elliptic curves and lattices, that we are simply not smart enough to discover - but bots soon will be?
This is a major part of the reason why for the past year ethereum's lean roadmap has been going in the "hash-only" direction: no lattices, no ML-DSA, no Falcon, no lattice-based commitments inside ZK proofs, etc. Signatures in lean ethereum are all hash-based, either WOTS or SPHINCS-.
For signatures and proofs, we already know how to go hash-only. The bigger challenge is for *public-key encryption* - and this goes far beyond blockchains. Secure communication, anonymizing protocols, lots of things need public-key encryption.
And unfortunately there are long-standing mathematical theorems showing why public-key encryption cannot be done with hashes alone. You have to have some kind of trapdoor object that has at least one form of usable "structure" - either group theory (incl. isogenies) or lattices or code-based or potentially in the future even more newfangled and spooky things (local mixing?). But for anything that has structure, you should assume that AI will make at least some progress in breaking that structure. Here, one reasonable inference is that if you want to make something plausibly long-term secure, multiply the key sizes by 10.
To me that's a very plausible world and something not at all extreme to predict. If AI will bring us 50 years of math in 2 years, then that 50 years of math may very plausibly include a "naive factoring -> GNFS" level of improvement to our ability to break lattices. In that world, lattices will still exist, but they will have to be significantly bigger to guarantee the same level of safety.
And at those new larger sizes, hash-based constructions will beat lattice-based constructions on concrete efficiency in every use case where hash-based constructions are possible at all.
Theoretically, of course it's possible that hashes are broken too (eg. P = NP would imply that). But I think P = NP is very unlikely. And intuitively, it's much more likely that a mathematical object has exactly no exploitable structure (like hashes are intended to), than that a mathematical object has exactly ~3 forms of exploitable structure (for elliptic curves: associativity, Schoof, pairings) and not some secret fourth form of structure we have not yet discovered that greatly degrades its security (for elliptic curves, ECDLP and pairing security). Similar for LWE, SVP, RLWE and the zoo of lattice problems.
For this reason, we do not yet see any reason to worry and start padding the byte size of hashes (if we start to worry more, we would pad the round count first before doing anything to the byte size).
Concrete TLDR, my own personal views:
* Hash-based > lattice-based, in those situations where hash-based is possible at all
* For anything lattice-based, be much more paranoid on param sizes. Remember that blockchains are only a small portion of the cryptography story; this point goes far beyond blockchains and applies to eg. access to websites, secure messaging, Tor / VPNs ...
* For privacy protocols, strongly favor NOT putting encrypted notes onchain. Instead, send them offchain through some third-party mechanism.
* If it's not difficult for you, keeping your funds in addresses which have not yet been used to make a transaction is a good idea. If it's easy for you, do it. **But be careful about migrations; I personally have lost more money in botched migrations than I have lost in all hacks combined**.
* For multisig wallets, doing confirmations offchain is better than onchain, because this way the signatures of signer wallets do not get exposed to the public, so if ECDSA falls to AI much faster than expected, at least the multisig "gracefully degrades" to a 1-of-1 where the 1 is whoever was gathering the signatures - a much better place to be than "anyone can take the money"
firefly.social/post/x/210783…
If you own crypto, this is genuinely scary.
A top bitcoin researcher, who actively works at the Ethereum foundation, is telling us to move our crypto to a bunker.
He thinks there is now a reasonable possibility that superhuman AI discovers a way to break the cryptography securing Bitcoin and Ethereum before quantum computers do.
That sounds insane, but there’s some important context.
Bitcoin and Ethereum rely heavily on ECDSA. At a very basic level, your wallet has a private key, and mathematics allows you to derive a public key from it.
Going forward is easy.
Going backward and figuring out the private key from the public information is considered effectively impossible.
A huge amount of crypto security rests on that assumption.
Drake is worried that AI could discover some completely new mathematical shortcut that changes it.
Why is he worried about this now?
AI has made some pretty ridiculous advances in mathematics this year.
Earlier this year, an OpenAI model autonomously disproved a famous Erdős conjecture dating back around 80 years, using mathematical techniques researchers hadn’t expected.
Then, literally the day before Drake made this post, OpenAI released hundreds of AI-generated mathematical results across hundreds of problem families.
One of those results broke through a longstanding assumption involving the complexity of integer multiplication.
That does NOT mean AI broke ECDSA.
It didn’t.
The signal is that AI is increasingly demonstrating an ability to find novel approaches to mathematical problems humans have studied for decades.
At the same time, researchers have been making progress on the other threat: quantum computers.
Estimates for the resources required to attack elliptic-curve cryptography have fallen significantly, and researchers, including Drake himself, have been working on making those attacks more efficient.
So his concern is basically:
What if everyone is preparing for quantum computers to eventually break ECDSA, but superhuman AI discovers a mathematical shortcut first?
That’s the part he thinks the crypto industry isn't sufficiently prepared for.
If ECDSA were efficiently broken, the consequences could obviously be catastrophic. Depending on the attack and what information an address has exposed, an attacker could potentially recover private keys and steal funds.
Drake isn't saying this has happened.
He isn't even saying it's definitely going to happen.
He's saying that recent AI progress has moved the probability from something he could comfortably ignore to something worth preparing for.
His recommendation is surprisingly simple.
If you're a major holder, exchange, custodian, ETF, treasury, etc., consider moving assets from heavily used addresses into fresh addresses that have never signed a transaction.
Then don't use those addresses.
That limits the cryptographic information you've exposed and potentially buys you additional protection if a vulnerability is suddenly discovered.
For the really large players, he thinks the industry should begin preparing to move beyond ECDSA entirely and toward post-quantum / hash-based cryptography.
The downside is enormous, the precautions are relatively cheap, and AI progress is moving fast enough that waiting until an attack exists may be too late.
He thinks it's now reasonable to consider the possibility that ECDSA could break before AGI/Q-Day, potentially on a timeline of months rather than years.
Nothing publicly known can currently do this.
But his argument is essentially:
For decades we've protected trillions of dollars with mathematical problems because humans couldn't figure out how to solve them efficiently.
We're now building machines that may become much better at mathematics than humans.
Maybe it's time to stop assuming those two facts can coexist forever.
Today I call upon the blockchain industry to calmly begin planning for "bunker mode". My personal recommendation is to set in motion a controlled mass migration of assets to fresh addresses, i.e. addresses whose pubkeys remain hidden behind a hash.
Holders, starting with large and sophisticated ones, should consider moving the bulk of their funds to addresses that have never signed a transaction. And when they do sign one, they should also move remaining funds to a new address (possibly generated from the same seed phrase).
Don't rush. While I believe there is cause for action a rushed migration would do more harm than good. Don't panic either. Moving assets to protected addresses is a simple, preventative step which does not require new cryptography or new wallets.
IMO it is now reasonable to brace for the possibility that ECDSA breaks before qday, in the worst case in months not years. By "break" I mean fast private key recovery (e.g. in one week) on available hardware (e.g. a large GPU cluster).
Recent days have been humbling for human mathematical intuition. Long-held, unquestioned hypotheses have fallen. This includes the n log(n) bound for integer multiplication and the 3SUM conjecture. In hindsight, May's unexpected disproof of the Erdős unit distance conjecture was our warning shot.
Yesterday's OpenAI drop made it clear that mathematical superintelligence is upon us. They say there are weeks where decades happen. We are about to live through weeks where centuries of mathematical progress happen. Could our magic 64-byte ECDSA signatures be too good to be true? Was it just security through obscurity all this time?
Elliptic curves feel especially vulnerable to superintelligence. Curves carry rich structure, with room for fancy tricks like Schoof, Frobenius, pairings. (By contrast, hashes are designed to minimise algebraic structure.)
Separately, as Ewin Tang can attest, an efficient quantum algorithm sometimes foreshadows an efficient classical one. We should be open to the possibility of a classical counterpart to Shor that breaks elliptic curves and RSA at once.
Also noteworthy is the striking under-representation of cryptographic breakthroughs among the 722 mathematical results OpenAI published. I've witnessed first-hand the US government censoring academic quantum cryptanalysis results. Backroom interventionism is my base case.
I urge large, sophisticated actors to lead by example. Project11's "risq list" (bitcoin-risq-list.projecteleven[.]com) is a great tracker of exposed BTC pubkeys. Binance, Bitbank, Robinhood, Bitfinex, and Tether have an opportunity to harden their cold storage. Next month I'll address institutions in London in a live Q&A (forum.ethereuminstitutional[.]org/london-2026).
Again, please do not rush. Wallets holding under 50 BTC enjoy partial cover from "Satoshi's shield", i.e. his 20K exposed addresses that hold 50 BTC each. Load-bearing signers like oracles and L2 security councils should consider rotating ECDSA pubkeys with every signed message and/or multi-signing with a hash-based schemes like SPHINCS.
Exiting bunker mode safely will require post-AI cryptography. My inclination is to go all-in on hash-based cryptography and avoid structured mathematical assumptions entirely, whether from curves, lattices, or isogenies. A single battle-tested hash (e.g. from the SHA or BLAKE families) yields plausible post-AI security.
The Ethereum roadmap on strawmap[.]org fully embraces hash-based cryptography with end-to-end formal verification as a response to the quantum threat. Those timelines must now be revisited and accelerated in light of mathematical superintelligence. I'll be pushing for maximum defensive acceleration.
Reposting as it’s more relevant than ever: many don’t have an opinion on something until they’re asked about it, at which point they cobble together a viewpoint from whim and half-remembered hearsay, before deciding that this 2-minute-old makeshift opinion will be their new hill to die on.
26
Getting offended by an opposing position doesn't make it wrong, just as feeling morally certain doesn't make it right!
The discipline of debate has the great benefit of forcing the debater to confront weaknesses and contradictions in their own thinking. That is increasingly rare, particularly when people treat even examining an opposing argument as a form of betrayal.
1
18
“The highest form of peace is having no desire to be seen, validated or understood by anyone.”
— Epictetus
30
If you find yourself losing patience with someone, remember that they are an ape adrift in an alien world, born into a struggle they did not choose, bullied by impulses they cannot control, searching for answers they will not find, and condemned to a fate they do not deserve.
2
22
What if the future already exists?
Physics has a strange idea called the Block Universe.
Past, present, and future all exist together.
Nothing is “becoming.” Everything simply is.
You’re just experiencing it one moment at a time.
Think of the universe not as something being created moment by moment, but as a four-dimensional structure containing every event in spacetime.
Imagine your life as a movie:
• Frame 1: You’re born.
• Frame 2: You’re 10.
• Frame 3: You’re 30.
• Frame 4: You’re 50.
• Frame 5: You die.
From inside the movie, you experience the frames sequentially.
You experience now. Then the next moment becomes now.
But from outside the movie, every frame already exists.
Maybe time isn’t something that moves.
Maybe we do.
1
23
A motte and bailey argument is a logical fallacy where a person mixes up two different positions: a bold, controversial claim and a safe, mild statement. Named after medieval castles, the "bailey" is the hard-to-defend courtyard, and the "motte" is the easy-to-defend hilltop tower. When people attack the risky bailey claim, the speaker retreats to the safe motte to pretend they only said something reasonable all along.
16













