AI that a person
can sign.
Five lines of research, eleven projects, and the evidence behind them.
Every figure on this site links to where it can be checked, with the date it was measured. What we have not measured is not here.
- 01 Documents that carry liability
Can a system turn messy operational documents into a record a named person can sign, and stop when it does not know?
- 02 Evidence and authority for machine decisions
With the same documents and the same retrieval budget, does a workflow that tracks evidence versions, policy, time and jurisdiction make fewer unsupported or inapplicable claims than a strong modern baseline?
- 03 Retrieval and memory for agent teams
How should a team of AI agents share memory so that a wrong note cannot outrank a reviewed fact?
- 04 The agent-readable web
What determines whether an AI agent can find, understand, consider and act on a business online?
- 05 Private, local AI
How much speech, search and assistant work can run inside the owner's perimeter, on hardware they control, without giving up quality?
of questions get the right passage ranked first by search with reranking. Plain text search manages 38%.
MEMORY-BENCH.md · 2026-11-11 ↗times a planted false note wins without trust tiers. With reviewed files preferred, it never does.
MEMORY-BENCH.md · 2026-11-11 ↗recall at rank one with the better of two rerankers on the Russian demo set. The other reaches 0.750.
COMPARISON.md · 2026-11-11 ↗A language model will answer anything.
In work that carries liability, an answer is worth something only when a person with authority can sign under it. That takes a source for every value, a rule for every decision, and a system that stops when it does not know. The lab studies where that line is, one domain at a time.
Five lines of research.
What exists today.
Trade-document engine
Incomplete supplier lines become a reviewable record with sources and gaps. A person keeps the decision.
Document cross-check
Reads a company's published documents against each other and shows every place where they differ, with the document, the date and the page.
Services-BGE
Local embedding and reranking services for hybrid search, and a shared, measured memory for AI agents.
Precedent
A person's correction to an AI judge becomes a rule that states its reason, is tested against every case already checked, is approved by a second person, and can be reverted.
VoiceToText
Private voice typing for Apple Silicon Macs. Hold a key, speak, let go, and the text appears at the cursor.
Agent Web Lab
Open experiments on how AI agents discover, interpret, compare and act on what a business publishes on the web.
Written for people and for their agents.
One source of facts. People get pages. Agents get the same facts as plain text and structured data.
Three ways in.
A research question
A scoped study with a written protocol, a baseline and findings, including the negative ones.
Bring a questionA pilot
One workflow, your documents, criteria agreed first. A working slice at the end.
Discuss a pilotPartners and investors
By introduction. Technology partners, licensing, strategic capital for a line of research.
private@scanovich.aiOne researcher. A governed team of agents.
Aleksandr Mordvinov
Founder · Principal researcher
More than fifteen years in work where documents carry liability: import and certification of electronics, regulatory permits, customs clearance. Since 2024 he has built the lab’s systems himself.
Each agent in the lab owns one zone. They share one memory, where reviewed facts outrank unreviewed notes. Nothing leaves in the lab’s name without his signature.