Your data can stay in Australia.
So can the AI.
Fixed-price Australian AI processing for the emails, documents, customer state, knowledge and operational workflows running through your connected business systems.
The sovereignty gap
Owning the database is only half the answer
Bringing your email, documents, CRM and operational data into one controlled database gives you a single source of truth and control over where it is stored.
But each time an overseas model summarises an email, evaluates a customer, classifies a document or runs an automation, the information required for that job leaves the environment for processing.
OpenAI and Google remain sensible choices for many workloads. They also introduce offshore processing, variable usage costs and another external dependency. Cheap open-source models do not solve that problem if they are still hosted overseas.
We provide the missing inference layer.
Part of the operating system
Ongoing business inference, not a document bureau
A batch archive can be processed during onboarding. The lasting value is what happens next: new information is processed as the business operates.
Email processing
Classify new communications, extract intent, identify entities and prepare operational context.
Document ingestion
Extract, clean, classify and index new files as they enter the connected knowledge system.
Customer state
Re-evaluate what is happening with a customer as messages, orders, tasks and meetings arrive.
Knowledge processing
Turn new operational information into searchable organisational knowledge and current summaries.
Scheduled agents
Run recurring analysis, reporting and operational workflows without sending routine work offshore.
Application inference
Power search, extraction, classification and drafting inside custom business applications.
01
Australian processing
Routine business inference runs on infrastructure operated in Australia.
02
Fixed monthly pricing
Normal operational processing uses a known monthly price rather than an unpredictable token bill.
03
Pinned open-weight models
Models are selected for the workload and versioned instead of silently changing underneath it.
04
Auditable execution
Processing can be logged by workload, model version and timestamp for operational visibility.
Model choice remains open
Australian by default. Frontier when the work requires it.
The platform remains model-agnostic. Australian open-weight models can handle routine, high-volume business processing while Claude, Gemini or OpenAI remain available for work where their capability provides a material advantage.
Routine operational work
Route recurring email, document, knowledge and customer-state processing through Australian inference.
Selected frontier work
Use an external frontier provider when the task needs it. The exact data-minimisation and PII-handling policy is defined during solution design.
Vibe coding and development
Claude Code and Codex CLI remain available for application development and technical work as a separate execution path.
Pricing
A known monthly number
Australian AI processing is independent from your platform and data-hosting choice.
Bitscape Australian AI
$750 AUD/month
First 5 seats included
- ✓ $150 AUD per additional seat
- ✓ Fixed monthly pricing
- ✓ Routine processing in Australia
- ✓ Independent of deployment choice
This is a separate product from the managed token packs — not a cheaper source of tokens. It runs on dedicated Australian hardware reserved for your workloads, which costs the same whether it is busy or idle. Sovereignty and control of the data path are the point; reduced cost on large batch workloads is a side effect.
Specialised high-volume processing and private infrastructure are scoped separately. All prices exclude GST.
Control the data path, not just the database
We can map which workloads stay in Australia, which require a frontier model and what information each execution path is allowed to receive.
Book a scoping call