sebbi.pro

Everyone is watching the system. Almost nobody is underneath it.

This is the whole stack, top to bottom. The tools most organisations rely on sit to the side and watch. The place a decision actually has to be proven is the layer beneath it — and that layer is nearly always empty.

bolted on · watches from the side
observability

Monitoring & dashboards

Logging platforms, dashboards, alerting. They read what the system emits and show it back to you. The record they keep lives in a database your own team can edit, so it says what you currently claim happened — not that nothing changed it since.

watches · cannot prove
bolted on · reads the output
guardrails

Filters & guardrails

Content filters and policy layers that inspect what a model says. Useful, but they act on the text after the model has produced it, and they keep no evidence a regulator can check without trusting the vendor who wrote them.

filters · cannot prove
↑ everything above watches after the fact ↑
where the decision happens

The execution layer

The model, the agent, the automated decision itself — the moment something is actually decided and acted on. This is the event that has to be evidenced. It is also the moment the watching layers above only ever see second-hand.

underneath the decision · seals it as it happens

The evidence layer — where sebbi.pro sits

Each decision is sealed into a hash chain at the moment it is made, anchored to a clock nobody controls, and cross-witnessed by independent systems. Not a record you keep and hope is believed — a record anyone can verify with your company switched off.

hash chainexternal anchorindependent witnessespublic verification
proves · cannot be edited

Why the watching layer can't carry autonomous decisions

When software did what it was told, watching it was enough — the inputs implied the outputs, and a log of the inputs was as good as a record of what happened. That is no longer true.

An autonomous system produces outputs you cannot derive by looking at the inputs. So the output has to be recorded as a fact in its own right, at the moment it happens, in a form nobody can quietly change afterwards. A layer that watches from the side cannot do that — by the time it sees the decision, the decision has already happened, and the only record is one the operator can edit.

This is why the volume problem bites. One reviewed decision a day can be watched by a person. Millions of automated decisions a month cannot — and the moment one is contested, "our dashboard showed it" is not evidence. It is an assertion with good formatting.

Four questions the watching layer answers "no" to

Can the people being audited edit the audit?

On the watching layer, yes — the record sits in a database they control. On the evidence layer, changing one record breaks every record after it.

Can it be checked with the vendor switched off?

On the watching layer, no — you log into the vendor to see it. On the evidence layer, a standalone verifier checks it with no account and no network call back.

Can you prove a record predates the complaint about it?

On the watching layer, the date comes from a field the system could set to anything. On the evidence layer, the timing is fixed by a clock nobody involved controls.

Can you prove the human approved before the machine acted?

On the watching layer, order is not recorded. On the evidence layer, the reviewer's decision is sealed before the machine's verdict is shown to them.

Don't take the diagram's word for it. Check the layer yourself.

Every claim on the evidence layer is verifiable right now, with no account, with our company switched off. Start with the live chain, or read the full architecture.

Read the whitepaper See the live chain
AI news
The GuardianCalifornia sues OpenAI over safety concerns after rogue agent hacking14mThe GuardianAlbanese signs on to Trump’s AI supply chain pact before leaders’ meeting at Quad summit54mThe VergeCalifornia sues OpenAI over rogue AI hacking incidents1hTechCrunchCalifornia state AG sues OpenAI over rogue agent incidents1hThe VergeApple is testing daily usage limits for more complex Siri AI interactions2hThe VergeOpenAI wants to stop deepfakes from overwhelming child abuse tiplines2hThe DecoderOpenAI's GPT-6 Astra tackles the Standard Model's toughest math problem on a lattice2hBBC NewsTrump backs return of ex-Nvidia boss after 'Bring Jensen Back' campaign launched2hTechCrunchTrump says ex-Nvidia CEO Jensen Huang’s affair with an alleged Chinese spy did not threaten national security2hArs TechnicaUS moves to cut China’s remote access to AI chips2hThe DecoderMoonshot AI pivots Kimi models from coding to science as benchmark gains flatten3hThe VergeNvidia’s board is reportedly ‘not considering’ hiring Jensen Huang back3hThe GuardianTrump backs calls for reinstatement of Jensen Huang as Nvidia CEO3hThe GuardianAnthropic recognises the risks to humanity – but is its governance really up to the job? | Nils Pratley3hTechCrunchSam Altman on Nvidia without Jensen Huang: the industry will be ‘fine’3hTechCrunchFBI says Artex AI open-source hacking tool used to breach networks in ‘under three hours’4hTechCrunchJordan Belfort’s new AI voice startup, Tone, wants to sell like the ‘Wolf of Wall Street’4hThe DecoderAlibaba's Qwen4-Max takes on Opus 5.5, GPT-6, and Kimi K3.1 Pro at much lower prices4hThe DecoderClaude now builds interactive websites inside canvas graphics, reviving the Flash era with real HTML5hThe GuardianAI models ‘weak and fallible’ on climate science, UN-backed study reports5hMIT Technology ReviewBuilding trustworthy AI by fencing it in, not predicting it5hThe DecoderAnthropic says the key to agent memory is knowing what to forget, not how much to store6hArs TechnicaJensen Huang allies build a "war room" to get him back at Nvidia7hThe VergeThe Nvidia scandal isn’t slowing down8hThe GuardianCalifornia sues OpenAI over safety concerns after rogue agent hacking14mThe GuardianAlbanese signs on to Trump’s AI supply chain pact before leaders’ meeting at Quad summit54mThe VergeCalifornia sues OpenAI over rogue AI hacking incidents1hTechCrunchCalifornia state AG sues OpenAI over rogue agent incidents1hThe VergeApple is testing daily usage limits for more complex Siri AI interactions2hThe VergeOpenAI wants to stop deepfakes from overwhelming child abuse tiplines2hThe DecoderOpenAI's GPT-6 Astra tackles the Standard Model's toughest math problem on a lattice2hBBC NewsTrump backs return of ex-Nvidia boss after 'Bring Jensen Back' campaign launched2hTechCrunchTrump says ex-Nvidia CEO Jensen Huang’s affair with an alleged Chinese spy did not threaten national security2hArs TechnicaUS moves to cut China’s remote access to AI chips2hThe DecoderMoonshot AI pivots Kimi models from coding to science as benchmark gains flatten3hThe VergeNvidia’s board is reportedly ‘not considering’ hiring Jensen Huang back3hThe GuardianTrump backs calls for reinstatement of Jensen Huang as Nvidia CEO3hThe GuardianAnthropic recognises the risks to humanity – but is its governance really up to the job? | Nils Pratley3hTechCrunchSam Altman on Nvidia without Jensen Huang: the industry will be ‘fine’3hTechCrunchFBI says Artex AI open-source hacking tool used to breach networks in ‘under three hours’4hTechCrunchJordan Belfort’s new AI voice startup, Tone, wants to sell like the ‘Wolf of Wall Street’4hThe DecoderAlibaba's Qwen4-Max takes on Opus 5.5, GPT-6, and Kimi K3.1 Pro at much lower prices4hThe DecoderClaude now builds interactive websites inside canvas graphics, reviving the Flash era with real HTML5hThe GuardianAI models ‘weak and fallible’ on climate science, UN-backed study reports5hMIT Technology ReviewBuilding trustworthy AI by fencing it in, not predicting it5hThe DecoderAnthropic says the key to agent memory is knowing what to forget, not how much to store6hArs TechnicaJensen Huang allies build a "war room" to get him back at Nvidia7hThe VergeThe Nvidia scandal isn’t slowing down8h