The multisig superapp. Multiple vendors, multiple chains, multiple dapps. Your group, your keys, your funds, your way. Operations & treasury focused.
Miami, FL
Joined August 2025
- Tweets159
- Following7
- Followers224
- Likes567
Did OpenAI just crack ECDSA & steal ETH to fund a transaction???
Absolutely not lmao. There are many, valid reasons to stay on alert for exploits.
So let’s try to muddle the self-custody waters anymore folks…
OpenAI did nothing of the sort. An LLM found a public seed used for testing & derived an address using BIP84.
That’s all. Nothing “hacky” nor an exploit.
🤖 Made with AI
ColdCard, Trezor, Ledger. $330M + sensitive data lost just in 2026...
For those asking if HWWs are still worth it? YES.
The frequency & ferocity of the exploits isn't unique to HWWs, crypto, or even finance; LLMs are simply a forcing function for new security standards.
The good news? The math is still intact & every company failed in different, but addressable ways.
The bad news? *Everything* around the math is significantly more vulnerable & we likely haven't seen the last or worst exploits.
Never doom, stay informed & protect your stack!
Our first podcast cover on our multisig survivability research! Ofc always happy to sit down w/ @ECBSJ
For the full breakdown and how @getcofund is positioning itself around this risk model:
youtube.com/watch?v=UJhQ0CNf…
Cofund Multisig retweeted
Very hot takes here but as I was mentioning to @stephanlivera, three things have become clear to me through iterating on @getcofund registration:
1. “Easy seed storage” is basically as solved as it’s going to get. Privy/Turnkey, services that Shamir keys w/ one shard tied to a normal login, are actually a solid default option for 90% of people
2. What matters for products therefore is the lowest-common denominator, or the default for self-custody. This is entirely up to the founders definition of “self-custody” but we made peace with^
3. Past a digestible default, what matters after is optionality, ~CAN~ I use fancier opinions (importing a seed, directing to a HHW)?
$240M in hardware wallet exploits *just in 2026* 😵💫
A dizzying amount by any measure! (w/o counting future rugs from leaks)
Undoubtedly a blow to self-custody, but it's worth inspecting each case independently because *every* attack was different.
Which solidifies our takeaway for the last few months...
The math (encryption, hashes, signatures) isn't broken - at least not yet.
But *everything* around it is x100 more vulnerable than before because it's x100 CHEAPER to do the research (ty LLMs).
Trust the math folks, nothing else!
> 3.3% you'll lose your private key in the next 12 months
✅ fact-check verified true by real Satoshiheads ✅
The aggregate probability of losing your key in any given year (assuming no check-ins) is the combined probability of three independent events: seed, person, place:
2.0% -> p(lose_seed)
1.0% -> p(lose_person)
0.3% -> p(lose_place)
3.3%
Oh God. Oh no. someone hide this madman’s tweet before our CEO sees this & adds “multi-scheme” to the Q4 roadmap 😭
Making our way through the conference circuit folks 🫡
Ping us to talk anything multisig-related!
Seed Phrases vs Key Sharding: Multisig's UX Trade-offs | Setzeus SLP781
@setzeus of @getcofund argues most people won't use seed phrases, weighs email-login key splitting against verifiability, and explains policy multisig.
Timestamps
00:00 — Intro
00:37 — From SetDev's L2 Work to Multisig
03:20 — After the Coldcard Hack: Why Self-Custody?
07:56 — Justin Drake's Bunker Mode and ECDSA
10:13 — Are Bitcoiners Bad at Building Products?
15:59 — Seed Phrases vs Privy-Style Key Sharding
20:25 — Trade-offs: Verifiability and Lock-In
25:35 — Recommending Self-Custody After Hacks
30:40 — Families, Fiduciaries, and Founders
34:25 — Spending Policies via Taproot Script Paths
37:24 — Limit: Bitcoin Spend Caps Not On-Chain
39:59 — Key Rotation Means Moving Funds
41:11 — Cofund Pricing and Fees
42:03 — Thesis: Multisig Becomes Joint Accounts
> are all keys equal in a multisig?
Nope! Not exactly. Fascinating work is happening on a new class of signatures that give "weight" to specific keys: threshold/weighted/hierarchical signatures.
The implementation with the most progress & promise is known as FROST; the later add-on, ROAST, helps make the setup more stall-proof/tamper-resistant.
ALT FROST (flexible round-optimized schnorr threshold) Signature
ALT ROAST (robust asynchronous schnorr threshold) Signatures
ALT Side-by-side comparison of how ROAST pairs to improve FROST
Succession except Logan Roy properly plans his inheritance with a multisig:
> Kendall opens will
> checks his key is valid
> receives inheritance
end
Now with a few days of reflection, we wanted to share our thoughts on the superb @BTCtreasuries conference put on by @ejuline @TimKotzman & @pete_rizzo_
Three main takeaways:
1. Institutions are both quietly diversifying *&* growing
2. The disconnect b/w devs & institutions is fairly large
3. Three teams think privacy can bridge these parties
Overall an exceptional event & I suspect we’ll be more involved next year :)
Delightfully surprised to see privacy on bitcoin becoming the next meta!
Pulse check, which have you heard of?
Any other projects worth inspecting?
@allocinitxyz @idealgroup @lxdev_
58%Alloc Init
33%Ideal Group
8%LX Dev
12 votes • Final resultsVery interesting statist perspective here…
Numerous instances of CBP Search Authority abuse to confiscate phones. Reasonable that it’ll evolve into a choke point for self-custody too.
1. Seed Risk (misplaced, device bricked, theft)
2. Person Risk (death, incapacitation)
3. Location Risk (fire, flood)
What other frequent loss-of-quorum / multisig risks are we missing here?
Unsurprisingly, family wealth preservation is one of the biggest use-cases we're seeing; interestingly, however, the worry is *bi-directional*:
1. Parents (dad) worried about recoverability & inheritance for offspring
2. Offspring (son) worried about parents falling victim to elderly fraud scams
It's been a great litmus UX test for our upcoming v2 👀
Does your current setup pass the family test?
Cofund Multisig retweeted
I reviewed the PR to add the multisig wizard to Bitcoin Core:
github.com/bitcoin/bitcoin/p…
> Families looking to cold store
> Founders operating & saving
> Funds optimizing for yield
Multisigs are a lot more powerful than we've been told folks (potential major update coming soon...)
