Question 1

What is agentic AI in enterprise process automation?

Accepted Answer

An agent reads a document, decides what it is, and takes the next step itself rather than waiting for a person to trigger each stage. In TeamSync the actions it may take are declared in business rules and enforced at the platform layer, so autonomy is bounded by configuration, not by prompt wording.

Question 2

How is an AI agent different from a traditional workflow rule?

Accepted Answer

A rule fires on a fixed condition you defined in advance. An agent classifies ambiguous input and chooses a path, then hands anything above your defined thresholds to a person. The rules do not disappear - they become the boundary the agent operates inside.

Question 3

Do AI agents make us a high-risk system under the EU AI Act?

Accepted Answer

That depends on your use case, not on the platform. Where a deployment does fall in scope, every agent action and refusal is anchored to the audit chain and human-oversight evidence is generated from it, so the required documentation is a query rather than a project.

Question 4

What happens when an AI agent gets something wrong?

Accepted Answer

Agent actions are versioned like any other change, so a document or workflow state can be restored to the point before the agent acted, and the audit chain records the inputs, the rule that fired and the outcome.

Question 5

How long does it take to deploy an AI agent workflow?

Accepted Answer

Agents are configured in the workflow designer from document types, metadata and rules you already have. Custom code is optional rather than required, so the constraint is usually how quickly your rules can be agreed, not build time.

Question 6

How does TeamSync keep AI answers permission-safe?

Accepted Answer

Every retrieval checks the asking user's access rights at the moment they ask — not at indexing time. If a user can't open a document, the AI can't quote it to them. Permissions are enforced at the platform layer, not bolted on.

Question 7

Does TeamSync replace our e-signature tool?

Accepted Answer

Yes — TeamSync Sign provides legally binding e-signatures (SES, AdES, and QES levels, eIDAS compliant) with a tamper-evident audit trail, built into the same platform as your documents and workflows.

Question 8

What is compliance reporting?

Accepted Answer

It is the work of producing evidence that your controls operated as documented over a period. Where the evidence lives in an append-only ledger, that reporting becomes a filtered export rather than a manual reconstruction from several systems.

Question 9

What should an audit trail record to satisfy an auditor?

Accepted Answer

Actor, action, timestamp and context for every view, edit, download, permission change, workflow step and AI action, in a record that cannot be edited after the fact. If any of those are missing, the auditor has to take a system's word for it.

Question 10

Is a database log enough to prove a record was not altered?

Accepted Answer

Not on its own - anyone with administrative rights can alter a database log. Entries here are cryptographically chained and anchored with blockchain-backed timestamps, so an alteration breaks the chain and an auditor can verify integrity independently.

Question 11

How long do we have to keep audit logs?

Accepted Answer

It depends on the regime - FINRA 17a-4, FDA 21 CFR Part 11, HIPAA and SOX all set different periods. Retention schedules run automatically per regulation and document type, with legal-hold override, so multiple regimes can apply to the same estate.

Question 12

Can compliance reporting be automated for more than one regulation at once?

Accepted Answer

Yes. Controls are mapped to SOC 2 Type II, ISO 27001, GDPR, HIPAA, PCI-DSS, GLBA, FINRA and NERC CIP from one interface, so one set of evidence answers several frameworks instead of being assembled separately for each.

Question 13

Which compliance standards does TeamSync support?

Accepted Answer

TeamSync ships with controls mapped to SOC 2 Type II, ISO 27001, GDPR, HIPAA, PCI-DSS, GLBA, FINRA, NERC CIP, and more depending on your industry. Blockchain-backed timestamps create tamper-evident audit trails regulators can verify independently.

Question 14

Can we manage retention and legal holds?

Accepted Answer

Retention schedules run automatically per regulation and document type — archive, flag for review, or crypto-shred on schedule, with full legal-hold override and a complete chain of custody.

Question 15

What is contract compliance tracking?

Accepted Answer

It is the ongoing check that both sides are actually doing what a signed contract requires - obligations met, deadlines hit, required language present. TeamSync extracts those dates and terms at signing and monitors them against the live record rather than a spreadsheet.

Question 16

Do we need separate CLM software if we already have a DMS?

Accepted Answer

Only if the two can stay in sync. A standalone CLM creates a second silo, so contracts live apart from the records they reference and each system carries its own audit trail. TeamSync runs the contract lifecycle on the same repository and the same trail.

Question 17

How do you track contract obligations after signature?

Accepted Answer

Renewal, notice-period and obligation dates are extracted at signing and drive alerts, review notifications and escalation rules ahead of each deadline, so an obligation surfaces before it lapses rather than after.

Question 18

How does AI find risky clauses across a contract portfolio?

Accepted Answer

Risk scanning reads the whole portfolio, not just new contracts, flagging indemnification, liability and missing-language issues and linking each flag to the compliance policy it breaches, so review runs continuously instead of at renewal.

Question 19

How much does contract lifecycle management software cost?

Accepted Answer

Pricing depends on contract volume and which capabilities you switch on. Because CLM runs on the same platform as your repository, signatures and audit trail, the comparison to make is against your combined tool spend rather than against a standalone CLM line item.

Question 20

Can TeamSync run fully on-premise or air-gapped?

Accepted Answer

Yes. TeamSync deploys entirely on your servers, behind your firewall, in your private cloud, or fully air-gapped with no internet connectivity. Your documents never pass through shared infrastructure or third-party servers.

Question 21

Is AI document summarization accurate enough for regulated work?

Accepted Answer

Accuracy is not the right test on its own - verifiability is. Every point in a TeamSync summary links to the passage it came from, so a reviewer confirms each claim against the source in one click rather than accepting the output at face value.

Question 22

Can AI summarize a contract without missing key clauses?

Accepted Answer

Long documents are processed in sections and then synthesised with structure preserved, and every point carries a citation, so a reviewer can see which clauses the summary drew on and check the ones it did not.

Question 23

Is it safe to use AI summarization on confidential documents?

Accepted Answer

Summaries are generated only from documents the requesting user is permitted to read, so there is no cross-permission bleed, and the whole pipeline can run inside your own environment including air-gapped deployments.

Question 24

How is AI summarization different from an extractive summary tool?

Accepted Answer

Extractive tools lift sentences verbatim and lose the connective reasoning. Generative summarisation restates across the whole document and across multiple documents at once - which is only safe when every restatement is traceable to a source, as it is here.

Question 25

Can it summarize a whole folder or matter at once?

Accepted Answer

Yes. Select a folder, a matter or a search result set and the output is a single synthesis across every file, with per-source attribution showing which document each finding came from.

Question 26

What is AI document search and how does it work?

Accepted Answer

AI document search interprets a plain-language question, retrieves the passages most relevant to it from your own files, and answers from those passages. DocuTalk names every source it used, so the answer is checkable rather than asserted.

Question 27

How do you audit what an enterprise AI search tool returns?

Accepted Answer

Every DocuTalk answer carries citations back to the exact paragraph, page and file. Every query and answer is also written to the tamper-evident audit ledger, so a reviewer can reconstruct what was asked, what was retrieved and what was returned.

Question 28

Does DocuTalk hallucinate and how would we know?

Accepted Answer

Answers are generated only from retrieved passages, and each claim links to its source. If a claim has no citation there is nothing behind it - which makes fabrication visible rather than something you have to take on trust.

Question 29

Can we use AI document search without our data leaving our network?

Accepted Answer

Yes. DocuTalk deploys entirely inside your environment, including fully air-gapped installations, with no document content, embeddings or AI requests leaving your infrastructure.

Question 30

How is DocuTalk different from a general AI chatbot?

Accepted Answer

A general chatbot answers from public training data and has no view of your permissions. DocuTalk answers only from your corpus, checks each user's access rights at the moment the question is asked, and cites its sources.

Question 31

What are eDiscovery managed services?

Accepted Answer

Traditionally, a third party collects, processes and hosts your data for a matter. TeamSync inverts that: collection, hold, review and production all run inside the repository that already holds the documents, so there is no export and no chain-of-custody gap.

Question 32

How much does eDiscovery cost per gigabyte?

Accepted Answer

Per-gigabyte pricing is an artefact of exporting data to a third-party platform. When collection, hosting and review happen inside your own repository, the cost that disappears is the processing and hosting spend - the oil-and-gas example on this page cut outside-counsel eDiscovery spend by 60%.

Question 33

What are the EDRM stages and which does TeamSync cover?

Accepted Answer

The Electronic Discovery Reference Model runs from information governance through identification, preservation, collection, processing, review and analysis to production. TeamSync covers preservation through production on the same governed repository, with governance already in place upstream.

Question 34

Can we run eDiscovery in-house instead of using outside counsel?

Accepted Answer

For collection, hold and first-pass review, yes - predictive coding produces defensibility metrics with transparent methodology opposing counsel can verify. Counsel still makes the legal calls; what moves in-house is the data handling and the cost attached to it.

Question 35

What happens to a legal hold when an employee leaves?

Accepted Answer

Held content is preserved regardless of retention schedules or account status, and the hold itself is logged with who applied it and when, so departure does not create a preservation gap.

Question 36

What is digital signature verification?

Accepted Answer

It is the cryptographic check that a signature belongs to the named signer and that the document has not changed since it was signed. TeamSync seals each signed document and anchors it to the audit chain, so any later modification is detectable.

Question 37

What is the difference between an electronic signature and a digital signature?

Accepted Answer

An electronic signature is any signing intent captured electronically. A digital signature is the cryptographic method that proves who signed and that nothing changed afterwards. TeamSync supports simple, advanced and qualified levels so the method matches the risk.

Question 38

Which signature level do we need - SES, AdES or QES?

Accepted Answer

Simple signatures suit low-risk internal approvals, advanced signatures prove signer identity, and qualified signatures carry eIDAS-recognised certificates for the highest legal standing. The level can be enforced automatically per document type and risk level.

Question 39

How do you verify a signature years after it was signed?

Accepted Answer

Blockchain-anchored timestamps and long-term validation keep signatures verifiable after the signing certificate expires, and even if the issuing certificate authority is no longer operating.

Question 40

Can external parties sign without a TeamSync account?

Accepted Answer

Yes. Counterparties sign through a secure, time-limited link. Access is rate-limited and every view and action is logged to the same audit trail as internal activity.

Question 41

How long does implementation take?

Accepted Answer

Typical deployments go live in 2–6 weeks depending on scale and integrations. Migration tooling handles bulk import from shared drives, legacy ECM systems, and email archives with automated classification on the way in.

Question 42

How do we get our existing documents in?

Accepted Answer

Folder-based agents watch existing repositories and automatically classify, tag, and route documents as they are ingested. AI-powered metadata extraction makes years of records searchable from day one.

Question 43

What is a document management and workflow system?

Accepted Answer

It is a single system that stores enterprise documents and records and routes them through the review, approval and retention steps that govern them. TeamSync adds federation, so documents stay in the systems they already live in while search, workflow and governance run across all of them.

Question 44

What is the difference between a document repository and an ECM?

Accepted Answer

A repository stores and versions files. An ECM adds classification, workflow, retention and audit on top. TeamSync behaves as an ECM layer over the repositories you already run, so you get the governance without replacing the storage underneath.

Question 45

Can TeamSync replace SharePoint as our document repository?

Accepted Answer

It does not have to. Connectors index SharePoint content in place, so search, AI and workflow work across it on day one. Teams that later want to consolidate can migrate gradually, with version history and author attribution preserved where the source system exposes them.

Question 46

How long does it take to set up a document management workflow?

Accepted Answer

Because federation removes the migration phase, most teams connect their first source system and run a live workflow inside the initial deployment window rather than after a multi-quarter data move. Document types, custom fields and validation rules are configured without a developer.

Question 47

What file types can an intelligent repository store and search?

Accepted Answer

Documents, PDFs, Office files, scanned images, audio, video and CAD files are indexed and searchable from one interface under the same permission model, with OCR applied to scans so paper archives are searchable alongside born-digital files.

Question 48

What is intelligent document processing (IDP)?

Accepted Answer

IDP is the pipeline that turns an unstructured document into structured data: capture, OCR or ICR, classification, field extraction and validation. The governed version adds a confidence score per field and an audit record of what was extracted and by whom.

Question 49

How do you get structured data out of a scanned PDF?

Accepted Answer

Extraction runs automatically on ingest: OCR reads the page, classification identifies the document type, and type-specific rules populate the fields you defined - invoice number, effective date, counterparty - as metadata on the document record.

Question 50

Can OCR handle poor-quality scans, faxes and old archives?

Accepted Answer

Accuracy varies with scan quality, which is why every extracted field carries a confidence score. Anything below your threshold routes to a review queue rather than entering the repository unverified - so a bad scan becomes a short review task, not a silent data error.

Question 51

Do we still need a data-entry team after automating extraction?

Accepted Answer

Usually a much smaller one, focused on the low-confidence exceptions rather than every document - the insurance example on this page processes 500,000 handwritten claim forms with a fraction of the original team.

Question 52

How is extracted metadata used for search and retention?

Accepted Answer

Extracted fields become searchable metadata on the document record and feed the retention schedule, so a document's own contents determine how it is found later and how long it is kept.

Question 53

What does pricing look like?

Accepted Answer

Pricing is based on deployment model, user count, and modules. Book a demo and we will scope it against your environment — no pressure, no spam.

Question 54

What is role-based access control (RBAC)?

Accepted Answer

RBAC grants permissions to roles rather than to individuals, and people inherit access from the roles they hold. Roles here compose from groups, document types, metadata values and workspace membership, down to the individual document and field.

Question 55

What is the difference between RBAC and ABAC?

Accepted Answer

RBAC decides from who someone is in the organisation. ABAC decides from attributes of the request - document type, classification, metadata values. TeamSync uses both, so a role sets the baseline and attributes narrow it for sensitive material.

Question 56

How do we prove least privilege to an auditor?

Accepted Answer

Every permission change is written to the tamper-evident ledger, so 'who had access on this date' is a query rather than a reconstruction. Quarterly access reviews generate from that permission history instead of being compiled by hand.

Question 57

What RPO and RTO can we expect?

Accepted Answer

Backups run continuously as incrementals with a configurable full-snapshot cadence. A single document version restores in seconds and a full workspace in minutes; the specific targets are set per environment during deployment.

Question 58

Who can restore a deleted document, and is the restore logged?

Accepted Answer

Restore rights are themselves a permission, so only roles you grant can perform one, and the restore is written to the same audit ledger as every other action. Restores can also be staged into an isolated environment for a drill that produces evidence without touching production.

Question 59

What is enterprise data governance?

Accepted Answer

It is the set of controls that determine who may see, change or remove information, and the evidence that those controls held. In practice it only works when the identity model, the audit trail and the storage layer are the same system rather than three that have to be reconciled.

Question 60

Can you run enterprise AI without sending data to the cloud?

Accepted Answer

Yes. Full-scale LLMs and agentic AI deploy directly inside your network, including air-gapped environments, so DocuTalk, semantic search and workflow automation run with no cloud dependency and no third-party access.

Question 61

What is the difference between on-premise, private cloud and air-gapped?

Accepted Answer

On-premise runs on your own servers. Private cloud runs on infrastructure dedicated to you. Air-gapped has no external network connectivity at all, so updates arrive as signed offline bundles verified against published checksums.

Question 62

Does on-premise deployment mean we lose the AI features?

Accepted Answer

No - that is the usual trade-off and it is the one this deployment model removes. Models run inside the same boundary as the rest of the platform, so inference happens locally with no external calls.

Question 63

Who holds the encryption keys?

Accepted Answer

You do. Bring your own KMS or HSM; data is encrypted at rest and in transit with keys you control and can rotate, and in a self-hosted deployment nothing leaves your network - not documents, not embeddings, not AI requests.

Question 64

What is enterprise search software?

Accepted Answer

Enterprise search indexes the systems your organisation already runs and returns results filtered by each user's access rights. The semantic version ranks by meaning rather than literal word overlap, so a query finds the right document even when the wording differs.

FAQS

FAQS