Swfte Sovereign Intelligence Platform
AI you can trust.Intelligence you control.
The Sovereign Intelligence Platform for building, governing and operating AI across your organisation, from infrastructure and data to models, agents, workflows and business solutions.
One platform.Capability and control.
Capability and control, in every layer.
01 Agent thread
Ask the source of truth.
A person, or an agent, asks a question. Before anything is read, access is checked against whoever is asking. The answer cites its sources and says how sure each fact is.
- AvailableCortex, the directory graph with its history, and the seven evidence statuses on facts.
- Designed forPer-caller access checks across documents and tickets, document and ticket connectors, answers inside Cortex, and the approval action.
Illustrative example with a fictional company. Not a real customer, not real data.
/askWho approves EU refunds over the threshold, and has that changed this quarter?
I will check what you are allowed to see, then answer from the enterprise graph and the policy library.
- Checked identity and accessyou can see 3 of 4 sourcesDesigned for
- Read the enterprise graph: owners, systems, historyowners and reporting linesDesigned for
- Pulled the policy documentverified · Refund policy v4Designed for
- Cross-checked ticket historycorroborated · 3 ticketsDesigned for
- Flagged one stale factstale · 41 daysAvailable
Answer
Refunds over the threshold for EU customers need approval from Finance Operations [1]. The policy owner is Maria Lind, Head of Finance Operations [2]. The threshold was added this quarter [3], and the last three refund tickets followed it [4]. The backup approver named in the policy, Tom Berg, left 41 days ago [5].
- [1]Refund policy v4 verified
- [2]Enterprise graph: reporting line observed
- [3]Policy version history inferred
- [4]Ticket history corroborated
- [5]Directory: account disabled stale
Sources checked
- DirectoryReadable
- Policy libraryReadable
- TicketingReadable
- HR recordsNo access
Why one fact is stale
The directory shows the backup approver’s account was disabled 41 days ago. The policy still names him.
AvailableThe backup approver needs replacing. Shall I ask Maria Lind to update the policy?
Designed for/askBefore I route ticket FIN-2231: who approves EU refunds over the threshold?
I will check what this agent’s identity may read, then answer from the source of truth.
- Checked the agent’s identity and scopescoped to the agent: 2 of 4 sourcesDesigned for
- Read the enterprise graph: owners, systems, historyroles only, names withheldDesigned for
- Pulled the policy documentverified · Refund policy v4Designed for
- Cross-checked ticket historyout of scope for this agentDesigned for
- Flagged one stale factstale · 41 daysAvailable
Answer
Refunds over the threshold for EU customers need approval from Finance Operations [1]. Route the ticket to the Finance Operations approver group [2]. The named backup approver is no longer active [3], so the group queue is the safe route.
- [1]Refund policy v4 verified
- [2]Enterprise graph: role, not person observed
- [3]Directory: account disabled stale
Sources checked
- DirectoryReadable
- Policy libraryReadable
- TicketingNo access
- HR recordsNo access
Why one fact is stale
The directory shows the backup approver’s account was disabled 41 days ago. The policy still names him.
AvailableRoute FIN-2231 to the Finance Operations approval queue?
Designed for01 Inference
Every model behind one door.
Connect puts one governed API in front of 50+ model providers. It routes each request, fails over when a provider is down and holds spending to the caps you set.
- Routing rules per workspaceAvailable
- Failover across providersAvailable
- Usage caps and audit eventsAvailable
- Open-weight models on your infrastructureDesigned for
[ 02.A ] Connect, models and providers. Frame from the Connect trailer
02 Compute
Run AI where your rules say.
Start on the connected cloud, with customer data in AWS eu-west-1. Dedicated, private, hybrid and air-gapped placements are designed for, and scoped with you.
- Connected cloud, EU regionAvailable
- Dedicated cloud and GPU capacityDesigned for
- Private, hybrid and air-gappedDesigned for
Illustration[ 02.B ] Compute capacity, illustrated. Not a product screen
03 Enterprise Intelligence
What your organisation knows, put to work.
The Enterprise Intelligence Platform: Cortex, the desktop AI workspace, and the company brain designed to sit behind it. Ask in plain language, see what changed, and build from what you find.
- Desktop AI workspaceAvailable
- Directory data into a graph you hostAvailable
- Analyse and visualiseDesigned for
- Build an agent from a findingAvailable
[ 02.C ] Cortex, the day in one view. Frame from the Cortex trailer
04 Studio
Describe the work. Studio builds it.
Governed agents and workflows, built visually or by describing them, bounded by your policies and handed to a person when it matters.
- Build by describingAvailable
- Policies attached to runsAvailable
- Hand-off to a personAvailable
[ 02.D ] Studio, a hand-off workflow. Frame from the Studio trailer
05 Governance
Approval before action. A record after.
Governance starts with Nexus, the trust fabric across every layer: identity, policy, approval and a run ledger. Compliance-by-design means the controls and evidence for your own compliance work.
- Approval gatesAvailable
- A run ledger for each runAvailable
- Identity for every agent and endpointDesigned for
- Trace from data to outcomeDesigned for
Illustration[ 02.E ] Nexus approvals, illustrated. Not a product screen
Product frames from Swfte trailers. Compute and Governance frames are illustrations, not product screens.
03 See it work
Ask once. Watch it happen.
Pick a product and a result. Each clip is the real product, cut from its trailer, except Nexus, which is shown as an illustration.
Product
Ask Studio
Build a voice agent that answers support calls, keeps private details out of the conversation and hands the call to a person when it should.
Ask Connect
Send support traffic to our preferred model, fall back when it is down, and stop spending at the monthly cap.
Ask Cortex
What did we decide in the pricing review, who owns the follow-ups, and can you turn it into a weekly check?
Ask Nexus
Before the procurement agent raises a purchase order above the threshold, ask the budget owner and record the decision.
IllustrationIllustration
Illustration[ 03.A ] Excerpts from the Studio, Connect and Cortex trailers. Muted, looped, captions on request.
[ 03.B ] Nexus: illustration, not a product screen
04 Agent thread
Describe a workflow. Watch it get built.
One sentence becomes a workflow with its systems, an approval gate and a policy, tested before it goes live.
- AvailableBuilding from a description, connecting systems, approval steps that pause a run, test runs and publishing.
- Designed forWriting the policy from the builder. Policy is enforced today on runs that have one attached.
Illustrative example with a fictional company. Not a real customer, not real data.
/new-workflowRoute refund requests over the threshold to finance for approval
I will draft it, connect your systems and test it before anything goes live.
- Drafted the workflow5 stepsAvailable
- Connected the systemsTicketing, Payments, EmailAvailable
- Added an approval gateFinance Operations, over thresholdAvailable
- Set the policyno card data in prompts, EU region onlyDesigned for
- Ran a test3 cases: 2 automatic, 1 to approvalAvailable
Workflow
Test run
- Case 1, under thresholdrefunded automatically
- Case 2, under thresholdrefunded automatically
- Case 3, over thresholdwaiting for Finance
Policy
No card data in prompts. Runs stay in the EU region. Refunds over the threshold wait for a person.
Designed forReady to publish. Refunds over the threshold will wait for Finance Operations.
Available05 Controlled autonomy
Choose how much you hand off.
Five levels, set per agent and per action. Move up when the evidence says you can. Levels 1 to 3 are available today; 4 and 5 are designed for.
L1 Level 1 of 5
Assist
AI recommends. A person decides and acts.
- The agent
- Drafts, summarises and recommends.
- Your people
- Decides and takes every action.
- Control
- Read access to approved data only.
- Status
- Available today, per agent and per action.
L2 Level 2 of 5
Approve
AI prepares the action. A person approves it first.
- The agent
- Prepares the action and asks.
- Your people
- Approves or rejects each action.
- Control
- Approval gate before anything changes.
- Status
- Available today, per agent and per action.
L3 Level 3 of 5
Supervise
AI acts within limits while people watch the record.
- The agent
- Acts inside the limits of its policy.
- Your people
- Reviews the run ledger and steps in.
- Control
- Policy at runtime, on runs that have a policy attached.
- Status
- Available today, per agent and per action.
L4 Level 4 of 5
Autonomous
AI acts on its own within strict policy and risk bounds.
- The agent
- Completes the job end to end.
- Your people
- Sets the bounds and audits the outcomes.
- Control
- Risk bounds, escalation and approval above thresholds.
- Status
- Designed for. The platform position, not yet a shipped setting.
L5 Level 5 of 5
Adaptive
AI improves how it works, inside controlled boundaries.
- The agent
- Learns from outcomes within its bounds.
- Your people
- Owns the boundaries and approves changes.
- Control
- Changes to behaviour stay inside the boundaries you set.
- Status
- Designed for. The platform position, not yet a shipped setting.
06 What we stand behind
Six things we will put in writing.
One platform, five ways in.
Inference, Compute, Enterprise Intelligence, Studio and Governance are entry points into one platform, not five tools to stitch together. Start with one and add the next when you need it.
Governance runs inside the work.
Policy changes what an agent can do while it runs, not after the fact in a report. Live today for runs that have a policy attached.
People approve what matters.
Agent actions can require a person to approve them first. You choose which actions, at which level of autonomy.
Every model behind one door.
One API in front of 50+ model providers, with routing, failover and usage caps, so a change of model is a setting, not a project.
Your infrastructure, your choice.
Connected today. Dedicated, private, hybrid and air-gapped placements are designed for and scoped with you.
Evidence, not adjectives.
Each run is recorded. We describe controls and evidence for your own compliance work and say plainly what we do not hold yet.
07 Where we are today
Plain facts, no badges.
What we hold, what is in preparation and what is not held. The trust centre carries the detail.
- SAML SSO and SCIM
- In development In development
- ISO 27001 and HIPAA
- Not held today Not held
- SOC 2
- Type I audit in preparation In preparation
- Customer data
- Stored in AWS eu-west-1 (Ireland) Available
[ 07.A ] As stated on /trust. Read the trust centre
What you probably want to know
Frequently Asked Questions
What is the Sovereign Intelligence Platform?
Swfte is the Sovereign Intelligence Platform: one platform to build, govern and operate AI across your organization, from infrastructure and data to models, agents, workflows and business solutions, with security, governance and control built into every layer.
How is Swfte different from other AI platforms?
Most platforms offer AI capability without control, or control without intelligence. Swfte is designed to give you both: governance runs inside the AI at runtime, so policy changes what an agent can actually do, and every action leaves evidence you can inspect.
Where does my AI run, and who controls it?
Swfte is designed so your organization decides where AI runs, which data it can see and what it is allowed to do, with options for dedicated cloud, private and hybrid deployment and data residency. Exact options depend on your plan and configuration.
Does Swfte hold security certifications?
Not yet. Swfte does not hold a SOC 2 report or an ISO 27001 certificate and does not sign HIPAA Business Associate Agreements today; a SOC 2 Type I audit is in preparation. Customer data is stored in AWS eu-west-1 (Ireland), encrypted at rest, and API traffic uses TLS 1.2 or later. Swfte is built compliance-by-design: it provides technical controls and evidence to support deployment within your applicable requirements, and the exact posture depends on your use case, jurisdiction and configuration. See swfte.com/trust for what is in place and what is in progress.
Can we start small and expand?
Yes. Most teams start with one API or one workflow and add models, private data, agents and dedicated infrastructure as they prove value. It is one platform at every step, so nothing is rebuilt as you scale.
How to build governed AI with Swfte
Swfte takes you from a first AI workload to governed, recorded execution in four steps, with no infrastructure to manage and no code required to start.
Start with an API or a workflow
Call any of 50+ model providers through one gateway, or describe a workflow in plain language in Studio. There is nothing to install.
Connect your data and set the boundaries
Ground agents in your company knowledge and approved tools, then define identity, permissions, data boundaries and policy for what they may do.
Choose an autonomy level
Begin with AI that recommends, require human approval where it matters, and raise autonomy one level at a time as the evidence supports it.
Deploy and keep the evidence
Deploy on Swfte cloud today; dedicated, private, hybrid and air-gapped placements are scoped with you. Each run is recorded, so you keep the evidence.
More from Swfte
Real work. Governed AI.
- Professional servicesBring the right answer into reach.Explore the capability
- Operations & logisticsGive every handoff a better starting point.Explore the capability
- TechnologyFind a clearer route to intelligence.Explore the capability
- Regulated operationsKeep consequential decisions accountable.Explore the capability
- Customer experienceMake room for a more considered response.Explore the capability
- People & enablementTurn a first day into a useful beginning.Explore the capability
Illustrative scenarios, not customer stories.
We see how work really happens: the chaos, the inefficiencies, the untapped potential. AI should amplify what makes us human, not replace it.
[ 10 ] One platform
Capability and control, in one platform.
Start with one layer and prove it. Add the next when you are ready. The controls come with you.