Biotech IP provenance: who can prove the discovery was yours?

The lab notebook was never built to win an argument

In biotech, the discovery is rarely the hard part. The hard part is proving, years later and in front of people who were not in the room, that the discovery was yours and that it existed when you say it did. A founder remembers the assay that worked at two in the morning. A patent examiner, an acquirer's diligence team, and an opposing counsel remember nothing. They read records. If those records can be backdated, reordered, or quietly edited, the science is sound but the provenance is worthless.

Conception and reduction to practice are legal events, not just scientific ones. The question that decides ownership is narrow and brutal. Who can show, with evidence a hostile reviewer cannot wave away, exactly what was known and exactly when. Paper notebooks, shared drives, and screenshots of instrument output answer that question badly, because every one of them can be altered after the fact and none of them carry their own proof of integrity.

Where ordinary records fail under pressure

Three failures recur, and each one has ended a claim that the science deserved to win.

• Backdating. A countersigned page proves a witness signed something. It does not prove the page was written on the date it carries, because both the entry and the date were produced by the same editable system.
• Silent revision. Cloud documents and electronic lab notebooks track versions, but an administrator with the right credentials can usually prune history. A diligence team cannot trust a timeline that the seller controls.
• Orphaned data. The raw instrument file, the analysis, and the conclusion live in three different places with no cryptographic link between them. Anyone can later swap the file that the conclusion supposedly rested on.

The common thread is custody. In every case the party that benefits from the record is also the party that can change it. That is precisely the conflict of interest a court, an examiner, or an acquirer is trained to distrust.

What a provenance record actually has to do

A record that survives scrutiny has to satisfy four conditions at once. It must fix content, so the bytes of the result cannot change without detection. It must fix time, so the moment of capture is independently verifiable. It must fix authorship, so the entry is bound to a specific person or instrument. And it must remain verifiable by someone who does not trust, and should not have to trust, the organisation that produced it.

Notarisation services and timestamping authorities address pieces of this, but they reintroduce the original problem in a new costume. You are once again trusting a third party to hold the proof, and that party can be subpoenaed, breached, discontinued, or simply located in a jurisdiction your counterparty distrusts. Sovereignty over the proof matters as much as the proof itself.

Sealing the record at the moment it is made

This is the tension Mickai is built to resolve, and it resolves it at the layer where most systems are weakest, the moment of capture. Mickai is a Sovereign Intelligence Operating System, a SIOS that runs fifty specialised AI brains on the operator's own hardware and stays fully capable offline. The research data never has to leave the building for the proof to be created.

Every consequential action inside the system is written to the Open Audit Record, the OAR. Each entry is sealed and signed with FIPS 204 ML-DSA-65, the published NIST post-quantum signature standard. Mickai did not invent that standard, it adopts it, which is the point. The signature is verifiable by anyone with the public parameters, not just by Mickai, so the proof does not depend on trusting the lab that produced it. Content, time, and authorship are bound together at the instant the result is recorded, and a later edit breaks the signature rather than hiding inside it.

For the long horizon that biotech demands, where a dispute can surface a decade after the bench work, the OAR does one more thing. Pantheon, Mickai's own sovereign Layer 1 with its native token PAN and a fixed supply of five billion, anchors a hash commitment of the record to Bitcoin. Only a fingerprint is anchored, never the underlying data, and no value is moved. Anchoring is not spending. The effect is a permanent, independent reference point that says this exact record existed by this date, checkable against the most widely replicated timestamp in the world.

Why this changes the diligence conversation

When provenance is sealed at capture, the burden of proof inverts in your favour. Instead of asking a diligence team to take your timeline on faith, you hand them records that verify themselves and an anchor they can check without your cooperation. The conversation moves from can we trust these notebooks to here is the cryptographic proof, confirm it yourself. That is a stronger position in a licensing negotiation, an acquisition, a patent prosecution, and an interference or derivation proceeding alike.

It also matters defensively. The same sealed trail that proves your priority is the trail that rebuts a claim that you took someone else's work, because it shows independent conception inside your own perimeter, on your own hardware, on a verifiable date. Trust Agent provides that perimeter. The provenance lives behind it rather than scattered across consumer cloud accounts that neither you nor a court can audit.

Provenance as a discipline, proven in practice

None of this is a thought experiment for the company that builds it. Mickai's own intellectual position rests on 101 filed UK patent applications, with around 2,234 claims, owned by Mickai LTD and naming Micky Irons as inventor. The standard the platform applies to a customer's discovery is the standard it holds itself to. Provenance is treated as a discipline that begins at the bench, not a document assembled in a hurry once a dispute has already started.

The discovery will always be the romantic part of the story. The provenance is the part that holds when the romance is over and the lawyers arrive. Build the proof into the moment of the work, seal it where it is made, and the question that decides ownership stops being a matter of memory and becomes a matter of record.

“Build the proof into the moment of the work. Then the question of who made the discovery stops being a matter of memory and becomes a matter of record.”