Contrary to everyone predicting that software hiring is going to disappear, I think we're about to see massive wave of hiring in software.
As Al makes it easier & faster to build systems, it'll also create entirely new attack surfaces.
We're going to see vulnerabilities & attack vectors that we've never dealt with before.
And the pace at which these systems are being deployed is arguably outpacing our ability to secure, monitor & maintain them.
We'll need more people building security infrastructure, managing Al agents, monitoring their behaviour & making sure the systems they're operating don't go rogue.
Al might reduce the effort required to build software, but it'll also increase the amount of software we need to secure & maintain.
Speaking at #Bluechip26 in Vienna🇦🇹, Nov 10-11.
Crypto safety is usually framed as hacks and exploits. For funds, a lot of the real risk is operational: fragmented positions, spreadsheet reconciliation, no live view of exposure.
Ping me if you are in Vienna 🫡
Guess, from next week on I can use my QRNG engine to generate a BIP39 mnemonic from quantum entropy, then load it onto a Trezor running debug firmware. This lets me test wallet generation with an independent entropy source.
LEDGER HARDWARE WALLET IMPLANTS WERE DISCOVERED OVER A YEAR BEFORE THIS WEEK’S SUPPLY-CHAIN ATTACK REVELATIONS
Renowned hardware security researcher and computer engineer Joe Grand (@joegrand), also known as “Kingpin,” explains how he was contacted in August 2025 by a man in Thailand who had discovered a compromised Ledger Nano X.
The man had purchased a cheap Ledger online to replace the battery in his existing device.
When he opened it, he discovered a smaller battery, unauthorized circuitry, additional wiring, and an antenna.
After posting his discovery on Reddit and reportedly receiving little interest from local police, he contacted Grand to investigate.
The discovery came more than a year before the Ledger supply-chain attack became a major public controversy, with nearly $100 million in cryptocurrency reportedly stolen this week.
The estimated loss from the Ledger Nano X grew meanwhile 102 mln USD. But what will be the consequences?
The Ledger implant shows why better supply chain controls and tamper-evident seals will never be enough. We need to rethink hardware wallets. No single device should ever generate, store, reconstruct or compute with a complete seed or private key. Not even temporarily in MCU registers, ALUs, memory or internal buses. Wallets should operate as quorums, using distributed key generation, FROST threshold signing and secure multiparty computation. Compromising one device must never be enough to steal funds. This also requires independent entropy, secure recovery, transaction verification and protection against firmware and side-channel attacks. I've started prototyping this with my older, self-built Trezors and an ESP32-P4 architecture combining NXP EdgeLock SE050 and Tropic Square TROPIC01 secure elements. The goal isn't an unhackable wallet. It's making a single compromised wallet insufficient to compromise self-custody. Call it mechanism design in hardware. We have to protect the possibilty for self custody.
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"
https://t.co/oVjwZog2lL
🏔️ The institutions have arrived
Robby was at CV Summit 2026 in Zurich, alongside CV VC, Ripple & Swiss Crypto Tax.
Key topics: institutional adoption, agentic AI, quantum threats & RWAs and the need for compliant tax infrastructure.
We’re building for the next cycle. 🚀
Disagree: Canada getting closer, UK regrets Brexit, Hungary is back. And US AI giants CEOs demand more regulation. Factcheck: it just takes a bit longer till we have full access to AI tools like Siri AI on iOS27. We will get it too.
I prefer to live in a strong EU than in a small nation state 🇪🇺
We are partnering with @HackenClub to integrate technical risk scoring into our stablecoin ratings.
Stablecoins can be fully backed and still depeg.
Reserves do not stop a contract from being drained, minting unbacked supply, or reading a stale price. This is technical risk.
Hacken is helping us score this risk by powering the I in SMIDGE: Implementation.
With this partnership, institutions can now use our ratings to assess factors like smart contract quality, supply integrity, oracle & bridge risk.
Bluechip is the first stablecoin rating agency to cover technical risk in an extensive manner as a full factor.
Updated ratings roll out gradually across all covered stablecoins over the next few months.
We are partnering with @HackenClub to integrate technical risk scoring into our stablecoin ratings.
Stablecoins can be fully backed and still depeg.
Reserves do not stop a contract from being drained, minting unbacked supply, or reading a stale price. This is technical risk.
Hacken is helping us score this risk by powering the I in SMIDGE: Implementation.
With this partnership, institutions can now use our ratings to assess factors like smart contract quality, supply integrity, oracle & bridge risk.
Bluechip is the first stablecoin rating agency to cover technical risk in an extensive manner as a full factor.
Updated ratings roll out gradually across all covered stablecoins over the next few months.
The problem isn't vaults. It's that we shipped the product layer and skipped the control layer.
In a fund: mandate, independent risk, custody separation, admin strikes NAV. Four functions, four parties.
In a vault: all four collapsed into the curator, and the only visible metric is trailing APR.
So the market prices yield and can't price risk. The cheap audits and the reaching for yield are downstream of that, not the cause.