Inside GEEQ: Stephanie So’s Layer-0 rethinks blockchain security

Security in blockchain is often framed at the implementation layers most people interact with: consensus designs, rollups, audits, and bridges. What is not discussed is beneath those debates: when a transaction first enters a system and is deemed worthy of being processed at all. 

For executives like Stephanie So, CEO and Founder of GEEQ, that distinction matters. Every business that touches digital assets, whether through payments, custody, or data pipelines, relies on assumed integrity at that first step. If that assumption is wrong, no compliance rule or patch can fix what follows.

“Consensus doesn’t guarantee honesty,” So said. “If everyone agrees on the same lie, you still get consensus, you just don’t get the truth.”

So’s critique is straightforward: many security conversations start after execution. Her response is to locate verification at the point of entry, before consensus forms. 

In GEEQ’s model, this Layer-0 stance is not pitched as a cure-all; it's an assumption that asks a narrower question: Can users and systems reliably tell when a validating node is honest, without relying on a majority?

Introducing Proof of Honesty (PoH)

According to the GEEQ whitepaper, this is precisely what Proof of Honesty (PoH) is designed to establish, making users “the ultimate arbiters” of whether a node’s ledger is correct. 

The paper describes PoH as providing (N-1)/N Byzantine fault tolerance, commonly illustrated as “~99% BFT” for typical network sizes, because a single honest node is sufficient for users to locate the canonical ledger.

For enterprise operators, PoH reframes reliability as a measurable property rather than a governance promise, an approach closer to financial auditing than to protocol voting.

The mechanism: detectable honesty and incentives

PoH runs alongside an incentive layer called the Catastrophic Dissent Mechanism (CDM). Each node independently evaluates the same candidate transactions; divergent results reveal dishonest behavior, and honest nodes can submit audits that seize a dishonest node’s bond. 

The whitepaper characterizes this as “Strategically Provable Security (SPS)”. For example, honest validation is the only coalition-proof equilibrium under the game-theoretic design.

“You don’t need everyone to behave,” So said. “You just need one node that refuses to cheat, and the math does the rest.”

The claim here isn’t that dishonesty disappears; it’s that detectable honesty becomes economically stable, and users have tools to prefer the honest ledger even under pressure.

GEEQ’s design keeps a narrow validation layer apart from an application layer. The validation chain handles $GEEQ accounts and basic transfers; the application chain can host business logic, tokens, or domain-specific data. This dual-ledger structure is presented as a way to reduce the validator attack surface while preserving flexibility for applications.

“All digital assets are data,” So said. “We keep separate ledgers so nothing gets lost; the asset and its description always travel together.”

Two practical implications follow from the paper’s description: one is that developers aren’t forced to put complex execution into the validation path, while the other is how provenance can be maintained without conflating executable code with core ledger security.

That separation is familiar to any CIO managing production and reporting databases: isolate the surface that can fail from the record that must endure.

The business model: paying for proofs, not just blocks

For operators who need verifiability at scale (e.g., telemetry, payments, audit trails), the whitepaper sets out two relevant constraints: low per-transaction fees, on the order of 1/100th of a cent, and horizontally scalable capacity via multiple interoperable chains. 

Those figures are the protocol’s stated target economics, not a guarantee of market pricing, but they explain why the team frames “proof calls” as a product surface.

“Every call to get proof is like a receipt,” So said. “And you can build businesses on that.”

The claim here is limited: if proofs are cheap and independently checkable, some compliance and assurance costs can shift from periodic audits to continuous verification. Whether that’s preferable will vary by use case and regulator.

So is not arguing that Layer-0 verification replaces existing security work. The narrower claim is that starting earlier, with user-verifiable honesty and incentives that punish divergence, can remove ambiguity that surfaces later. 

The whitepaper explicitly positions PoH as making users “the ultimate judges of truth,” with the application layer left free to evolve for specific needs.

“We’re not trying to be the fastest,” she said. “We’re trying to be the most honest.”

That’s a testable idea: not a forecast, not a guarantee, an architectural choice that can be evaluated on its own terms.

For boards weighing blockchain integrations, the question So raises is not speed or cost, but evidentiary certainty: can the system prove it behaved as intended?