Calc retweeted
ZTARKNEAR is back.
Join @zkDragon, @ilblackdragon, and @EliBenSasson to discuss the quantum threat and how Zcash, NEAR, and Starknet are preparing for it.
Tomorrow at 6 PM CET.
Join @zkDragon, @ilblackdragon, and @EliBenSasson to discuss the quantum threat and how Zcash, NEAR, and Starknet are preparing for it.
Tomorrow at 6 PM CET.
11
10
53
4,523
its been like 3 months of anthropic dominance
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.
They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
0
0
14
881
Calc retweeted
I’m playing the new @LootRealms Frontier - Devcon Edition!
New to Realms games, but I’m having a blast, and it’s completely free.
Even cooler: I’m nowhere near the top of the leaderboard, yet I grabbed 1,200 $LORDS (~$8) just from exploring the map.
New to Realms games, but I’m having a blast, and it’s completely free.
Even cooler: I’m nowhere near the top of the leaderboard, yet I grabbed 1,200 $LORDS (~$8) just from exploring the map.
6
7
33
678
Calc retweeted
The post-quantum path @Starknet is taking is revealing more serious loopholes in blockchain design than I thought.
StarkWare built a framework from NIST's 'general crypto-agility guidance' to test how agile blockchains are, and only one chain scored 8/8.
Bitcoin itself passes for second among the 10 chains scored, yet it faces one of the hardest migrations.
NIST wrote the guidance but doesn't rate chains, so StarkWare built the 8 questions and scored the networks itself.
It's a bit technical, but let's review it.
➥ Based on NIST's guidance on crypto agility, the 8 questions ask whether a chain can replace its cryptography and keep running:
1. Can an existing account switch to a new signature without a network upgrade?
2. Can its formats handle bigger keys and signatures?
3. Can one cryptographic part be replaced without redesigning the rest?
4. Can old and new signing run side by side undedr migration?
5. Has the transition been tested?
6. Can an attacker be stopped from forcing a weaker algorithm?
7. Can the old algorithm be switched off after migration?
8. Can existing data be migrated?
A point only counts if the path works today under current rules. Anything that needs a future upgrade scores zero.
Across 10 networks the scores are @Starknet 8, @Bitcoin 7, @Algorand 7, @NEARProtocol 7, @QRLedger 6, @Zcash 6, @ethereum 3, @solana 2, @Optimism 2 and @RobinhoodCrypto 1.
➥ The test that separates Starknet from the 7s is the first one.
On Starknet every account is a smart contract, so a wallet can replace its signature check and keep its address and funds. StarkWare confirmed this with @ready_co.
A coin on the Bitcoin network has to move to a new address to use new cryptography, and NEAR and Algorand accounts can only choose schemes the protocol has already switched on, nothing more.
What I respect most is what the score does not claim. It is not a security rating.
QRLedger scores 6 yet shows lower exposure on the same Hub, because its signatures are already hash-based.
Starknet's standard accounts still sign with elliptic curves, and its post-quantum Falcon-512 account is an unaudited mainnet demo that ordinary users can't adopt yet.
So my takeaway is that being able to change and having already changed are two different things, and a chain needs both.
Next, I'll break down where Starknet stands on the second. For now, what's your take on this?
StarkWare built a framework from NIST's 'general crypto-agility guidance' to test how agile blockchains are, and only one chain scored 8/8.
Bitcoin itself passes for second among the 10 chains scored, yet it faces one of the hardest migrations.
NIST wrote the guidance but doesn't rate chains, so StarkWare built the 8 questions and scored the networks itself.
It's a bit technical, but let's review it.
➥ Based on NIST's guidance on crypto agility, the 8 questions ask whether a chain can replace its cryptography and keep running:
1. Can an existing account switch to a new signature without a network upgrade?
2. Can its formats handle bigger keys and signatures?
3. Can one cryptographic part be replaced without redesigning the rest?
4. Can old and new signing run side by side undedr migration?
5. Has the transition been tested?
6. Can an attacker be stopped from forcing a weaker algorithm?
7. Can the old algorithm be switched off after migration?
8. Can existing data be migrated?
A point only counts if the path works today under current rules. Anything that needs a future upgrade scores zero.
Across 10 networks the scores are @Starknet 8, @Bitcoin 7, @Algorand 7, @NEARProtocol 7, @QRLedger 6, @Zcash 6, @ethereum 3, @solana 2, @Optimism 2 and @RobinhoodCrypto 1.
➥ The test that separates Starknet from the 7s is the first one.
On Starknet every account is a smart contract, so a wallet can replace its signature check and keep its address and funds. StarkWare confirmed this with @ready_co.
A coin on the Bitcoin network has to move to a new address to use new cryptography, and NEAR and Algorand accounts can only choose schemes the protocol has already switched on, nothing more.
What I respect most is what the score does not claim. It is not a security rating.
QRLedger scores 6 yet shows lower exposure on the same Hub, because its signatures are already hash-based.
Starknet's standard accounts still sign with elliptic curves, and its post-quantum Falcon-512 account is an unaudited mainnet demo that ordinary users can't adopt yet.
So my takeaway is that being able to change and having already changed are two different things, and a chain needs both.
Next, I'll break down where Starknet stands on the second. For now, what's your take on this?
1/ Bitcoin passes 7 of our 8 cryptographic-agility checks.
So why does it still face one of the hardest post-quantum migrations?
Because upgrading the network's cryptography is one thing. Moving everyone's BTC is another 🧵
25
10
140
14,582
Calc retweeted
Anthropic is using cryptography StarkWare’s founders co-invented to prove which AI model answered you.
With FlashProve, you can verify that a specific model produced an output while keeping its weights private.
The report cites the original STARK paper, highlights the use of d references formal verification work on Cairo and the S-two AIR. StarkWare co-founder Alessandro Chiesa is one of the report’s co-authors.
Another big validation for our Proving System.
The same cryptographic ideas that help scale blockchains are now helping make AI inference verifiable.
With FlashProve, you can verify that a specific model produced an output while keeping its weights private.
The report cites the original STARK paper, highlights the use of d references formal verification work on Cairo and the S-two AIR. StarkWare co-founder Alessandro Chiesa is one of the report’s co-authors.
Another big validation for our Proving System.
The same cryptographic ideas that help scale blockchains are now helping make AI inference verifiable.
30
43
196
16,924
Calc retweeted
Anthropic @AnthropicAI ❤️ zkSTARKs:
In collaboration with @StarkWareLtd co-founder Prof. Alessandro Chiesa and other brilliant researchers, Anthropic just open sourced a zk STARK system for proving the integrity of LLM inferencing, bringing us closer to verifiable AI.
As we said long ago (since 2018), zkSTARKs are the end-game: highest scale, best security (post-quantum). Now, also AI agile and supporting AI integrity.
anthropic.com/research/prova…
In collaboration with @StarkWareLtd co-founder Prof. Alessandro Chiesa and other brilliant researchers, Anthropic just open sourced a zk STARK system for proving the integrity of LLM inferencing, bringing us closer to verifiable AI.
As we said long ago (since 2018), zkSTARKs are the end-game: highest scale, best security (post-quantum). Now, also AI agile and supporting AI integrity.
anthropic.com/research/prova…
59
123
815
55,699
> The AI labs themselves are using @Starknet tech and formal verification to enable trust in their models
Everyone will use our tech and be happy
Everyone will use our tech and be happy
3
2
35
694
Subtle Starknet bullpost
- FV & Open source
We have spent several years pushing the most robust level of fv in the industry, verifying most of the Starknet stack.
That covers the prover encoding the Cairo VM, core CairoZero functions, the compiler’s building blocks, and the STRK20 pool, all including no double-spends.
All open source.
See more: starkware.co/blog/strk20-fo…
- distributed
All VC unlocks end in march 2027 🫡
0
0
2
46
Calc retweeted
The challenge: we prepare for the threats we know are already here. We prepare for the ones we know are coming. But what do we do about threats we can't anticipate just yet?
The answer: We maintain crypto agility.
We have to build systems that can adapt quickly and easily, even to threats we haven't anticipated yet.
Good article. Read 👇
The answer: We maintain crypto agility.
We have to build systems that can adapt quickly and easily, even to threats we haven't anticipated yet.
Good article. Read 👇
8
14
73
2,812
Calc retweeted
It’s very easy to setup @bot and make your agent earn.
Just ask it
“go make me money on the taskmarket”
It will find it (because we dominate SEO) install the wallet and attempt tasks. Then just ask it to monitor and complete tasks.
Just ask it
“go make me money on the taskmarket”
It will find it (because we dominate SEO) install the wallet and attempt tasks. Then just ask it to monitor and complete tasks.
11
7
52
4,438
Subtle Starknet bullpost
- FV & Open source
We have spent several years pushing the most robust level of fv in the industry, verifying most of the Starknet stack.
That covers the prover encoding the Cairo VM, core CairoZero functions, the compiler’s building blocks, and the STRK20 pool, all including no double-spends.
All open source.
See more: starkware.co/blog/strk20-fo…
- distributed
All VC unlocks end in march 2027 🫡
- FV & Open source
We have spent several years pushing the most robust level of fv in the industry, verifying most of the Starknet stack.
That covers the prover encoding the Cairo VM, core CairoZero functions, the compiler’s building blocks, and the STRK20 pool, all including no double-spends.
All open source.
See more: starkware.co/blog/strk20-fo…
- distributed
All VC unlocks end in march 2027 🫡
People shortly will learn that the only safe havens are
- open source
- formally verified
- distributed
Zillions
3
5
33
1,024
