AI Tools with Australian Data Centres: The Complete List

If you've started asking which AI tools actually store data in Australia, you're asking the right question. Most AI tools were built in the US and default to US servers unless you specifically choose otherwise, and in many cases pay more to do so. This guide lists the major platforms that do offer Australian data centre options, what tier or plan you need to access them, and where the boundaries of that commitment sit.

Data residency matters under Australian law because of Australian Privacy Principle 8 (APP 8), which covers what happens when personal information is disclosed to an overseas recipient. If your AI tool stores or processes personal information on servers outside Australia, you carry additional legal obligations under the Privacy Act 1988. For a plain-English explanation of why this matters for your business, see our guide on AI and the Privacy Act in Australia. APP 8 is one of thirteen Australian Privacy Principles that govern how businesses handle personal information, and our plain-English guide to the Australian Privacy Principles and AI tools covers what each principle means in practice.

This article is organised by category: productivity and document AI, CRM and business AI, communication AI, and storage and cloud AI. For each platform, this guide covers what Australian data options exist, which plan tier is required, and what the commitment does and does not cover. Tools are listed only where a genuine Australian data centre option exists. Tools with no Australian data residency option are not listed here.

Productivity and Document AI

Microsoft 365 Copilot

Microsoft operates two Australian data centre regions: Australia East (New South Wales) and Australia Southeast (Victoria). For Microsoft 365 business customers, core data for services including Exchange, SharePoint, OneDrive, and Teams is stored in the Australian region by default for tenants with a billing address in Australia. Microsoft publishes a data residency map and a Data Residency Advanced commitment as an add-on product.

Microsoft 365 Copilot is available on the Microsoft 365 Business Standard, Business Premium, and enterprise E3/E5 plans. The Copilot add-on lists at USD $30 per user per month, which works out to approximately AUD $46-47 per user per month (as of July 2026), on top of your base Microsoft 365 subscription, or around AU$560 a month for a 12-person team, close to the cost of a full day of a bookkeeper's time each month. Copilot operates within your existing Microsoft 365 tenant boundary, which means it inherits the data residency of your Microsoft 365 environment. If your tenant is already provisioned to the Australian region, Copilot processes and stores data within that same region.

What the commitment covers: data at rest in the Australian region for core Microsoft 365 services and Copilot interactions stored in that environment. What it does not cover: all AI model inference may route through Microsoft's global AI infrastructure, and the Data Residency Advanced add-on (approximately AU$3.30 per user per month, or around AU$40 a month for a 12-person team) is required if you need a contractual guarantee that data stays in Australia rather than a default placement. Review Microsoft's Product Terms and the Microsoft Trust Centre before relying on defaults for regulated data.

Take a 12-person bookkeeping firm in Melbourne that handles tax file numbers and superannuation details for two hundred clients. Before checking Australian data residency, the office manager assumed every tool the team used sat wherever it sat, and gave it no more thought than the office printer. After working through a list like this one, she moved the firm's document AI onto a Microsoft 365 tenant provisioned to the Australian region, and can now tell a client in one sentence exactly which country their file lives in.

Google Workspace (Gemini AI features)

Google Cloud operates multiple Australian regions, including Sydney (australia-southeast1) and Melbourne (australia-southeast2). For Google Workspace business customers, Google offers a Data Regions policy that lets administrators pin certain data types to a specific geographic region, including Australia. This is available on Google Workspace Business Standard, Business Plus, Enterprise Standard, and Enterprise Plus plans.

Google Workspace pricing starts at approximately AU$17 per user per month for Business Standard (as of June 2026), or around AU$204 a month for a 12-person team, less than most firms spend on printer paper and toner over the same period. The Data Regions feature, which enables Australian data pinning, is included from Business Standard upwards. The AI features in Google Workspace, branded under the Gemini umbrella, operate within your Workspace environment and are subject to Google Workspace's data processing terms rather than consumer Google account terms.

What the commitment covers: data at rest for covered Workspace data types when the Data Regions policy is set to Australia. What it does not cover: not all data types are covered by the Data Regions policy. AI model inference and some metadata processing may occur outside the pinned region. Google publishes a list of which data is and is not covered by the Data Regions policy in its Workspace Admin Help documentation. For business owners storing health information or financial data, read that list carefully before treating the Data Regions setting as a complete solution.

CRM and Business AI

Salesforce (Einstein AI)

Salesforce operates Australian data centres and offers Australian data residency as part of its Government Cloud and specific enterprise configurations. For standard Salesforce CRM customers in Australia, data is hosted in Australian infrastructure by default for tenants provisioned to the Australian region. Salesforce Einstein AI features are available on Sales Cloud and Service Cloud plans, starting from the Enterprise tier (approximately AU$220 per user per month as of June 2026, or around AU$2,640 a month for a 12-person sales team).

What the commitment covers: primary data at rest for your Salesforce org when provisioned to the Australian region. Einstein AI features operate within your org environment and process data within Salesforce's infrastructure. What it does not cover: some AI processing, particularly generative AI features introduced in 2023-2025, may route through Salesforce's global AI infrastructure partners. Confirm with Salesforce directly if you are in a regulated industry such as financial services, healthcare, or legal, and ask specifically about where Einstein generative AI model inference occurs.

HubSpot (AI features)

HubSpot does not currently offer Australian data residency. Its servers are based in the United States. HubSpot AI features, including its AI content assistant and AI-powered CRM tools, operate on US infrastructure. For Australian businesses using HubSpot with personal information from Australian customers, this means you have cross-border disclosure obligations under APP 8. HubSpot is included here as a note, not as a compliant option, because it is commonly evaluated alongside Salesforce in Australian SMB contexts. For the full data handling picture on HubSpot and CRM tools, refer to our AI data residency comparison table.

Communication AI

Microsoft Teams (with Copilot)

Microsoft Teams data residency follows the Microsoft 365 tenant configuration described in the Copilot section above. Teams chat, meeting recordings, and call data are stored in the region assigned to your Microsoft 365 tenant. For Australian tenants, this defaults to the Australian region. Teams Copilot, which summarises meetings and drafts messages, inherits that same regional storage for the content it generates and stores within your tenant.

Teams Copilot requires the Microsoft 365 Copilot add-on. It is not available as a standalone product. The data residency story for Teams is more mature than for some other tools because Teams has been offered to regulated Australian industries, including government, for several years. Microsoft publishes specific Teams data residency documentation separately from its general Microsoft 365 data residency pages, so check both if Teams is a core part of your business communications.

Zoom (AI Companion)

Zoom operates data centres in Australia and offers an Australian data residency option for Zoom Phone and Zoom Meetings through its Business Associate Agreements and enterprise data residency configurations. Zoom AI Companion, which provides meeting summaries and chat assistance, is included free on Zoom One Business plans and above (approximately AU$27 per user per month as of June 2026, which works out to around AU$324 a month for a 12-person team).

What the commitment covers: Zoom's Australian data centre option stores meeting recordings and certain account data in Australia when the configuration is set. What it does not cover: Zoom AI Companion's AI model processing uses third-party AI infrastructure, and Zoom's own data residency pages note that AI-generated content such as meeting summaries may be processed outside your selected data residency region. If AI meeting summaries contain personal information about clients or patients, this is a relevant gap to confirm with Zoom before relying on data residency settings alone.

Storage and Cloud AI

Amazon Web Services (AWS) and Claude

AWS operates the Asia Pacific (Sydney) region (ap-southeast-2) and the Asia Pacific (Melbourne) region (ap-southeast-4). For businesses building applications on AWS, data stored in these regions stays in Australia subject to standard AWS data processing terms. AWS does not move data between regions without customer instruction.

Claude, the AI model developed by Anthropic, is available through AWS Bedrock. AWS Bedrock is the managed AI service that lets businesses access large language models, including Claude, through AWS infrastructure. When you access Claude through AWS Bedrock in the Sydney or Melbourne region, your prompts, inputs, and outputs are processed within the AWS Australian region infrastructure rather than routed to Anthropic's own servers. AWS Bedrock is priced per token (input and output), with Claude model pricing starting from approximately US$0.003 per 1,000 input tokens for lighter models (as of June 2026). There is no flat subscription; costs scale with usage.

What the commitment covers: data processed through AWS Bedrock in an Australian region is handled within that region's infrastructure, subject to AWS's data processing agreements. AWS's enterprise agreement and DPA can be configured to support regulated Australian data requirements. What it does not cover: using Claude directly at claude.ai (Anthropic's own product) does not currently offer Australian data residency. The Australian data centre option for Claude is specifically through AWS Bedrock, not through Anthropic's direct consumer or API products. For businesses where Claude is part of a compliance strategy, this distinction matters.

pCloud (Cloud Storage with AI features)

pCloud is a cloud storage provider that offers a European Union server location option, giving users a choice of where their data is stored rather than defaulting to US infrastructure. pCloud is not an AI tool in the same category as Copilot or AWS Bedrock, but it is a compliant cloud storage option for Australian businesses that want to avoid US-default cloud storage when storing business documents, client files, and materials that AI tools will later process.

pCloud offers a lifetime plan (a one-time payment rather than a subscription), which is uncommon in cloud storage. The 2TB lifetime plan is priced at approximately AU$595 as a one-time purchase (as of June 2026), with EU server selection available as a paid add-on. The EU server option is not Australian data residency, but it does avoid US server default, which may be relevant if your concern is keeping data outside US jurisdiction rather than keeping it specifically in Australia.

pCloud is worth noting for businesses evaluating compliant storage options that integrate with their AI workflows. It is an affiliate partner of Need to Know AI, disclosed here in line with our affiliate disclosure policy. The recommendation is based on its genuine utility as a non-US-default storage option with a lifetime pricing model that reduces ongoing subscription costs. It is not a substitute for Australian data residency where that is a specific regulatory requirement.

Microsoft Azure (for businesses building on cloud)

Microsoft Azure's Australian regions (Australia East and Australia Southeast) have been available since 2014 and are among the most mature cloud regions in the country. For businesses running applications on Azure, data is stored in the Australian region by default when the region is selected at provisioning. Azure AI services, including Azure OpenAI Service, can be deployed within the Australian region, subject to service availability in that region at the time of deployment.

Azure OpenAI Service availability in the Australian region has expanded over 2024-2025 as Microsoft has added capacity. Not all model versions are available in all regions at the same time. Businesses with strict data residency requirements should check Azure's regional service availability page before architecting a solution that assumes all AI model versions will be available in the Australian region.

Azure pricing varies by service and consumption. There is no flat SMB subscription; costs are usage-based across compute, storage, and AI API calls. For a small business without cloud engineering resources, Azure is typically accessed through a Microsoft partner or managed service provider rather than directly. The data residency benefit of Azure applies to infrastructure built on it, not to end-user Microsoft 365 products, which have their own data residency story covered in the Copilot section above.

What These Commitments Do Not Cover

Australian data centre options from the platforms above generally cover data at rest. That means the files, documents, chat logs, and records stored in those platforms sit on servers in Australia. What most commitments do not cover, or cover only partially, is data in transit (data moving between your devices and the servers), AI model inference (the computation that happens when the AI processes your input), and metadata such as usage logs, account information, and timestamps.

This distinction matters for businesses in regulated industries. A law firm storing client documents in a Microsoft 365 Australian tenant is covered for the documents at rest. But if those documents are submitted to a Copilot prompt and the inference processing routes through a non-Australian AI data centre, the data in transit and the inference processing may fall outside the residency commitment. For professional services, healthcare, and financial services businesses with specific regulatory obligations, this gap needs to be confirmed with the vendor in writing before relying on data residency settings as your compliance answer.

For a full side-by-side comparison of how each tool handles data in transit, inference processing, and metadata, see our AI data residency Australia comparison table. For the specific question of what ChatGPT does with your data, see our guide on ChatGPT and customer data in Australia.

What to Do Before Choosing a Tool

The first step is to identify what type of data your business processes through AI tools. If it is internal, non-personal data, such as internal documents, marketing copy, or operational templates, the data residency question is less pressing. If it includes personal information about clients, patients, employees, or customers, you have Privacy Act obligations that make the data location question material.

The second step is to read the vendor's actual data processing agreement, not just the marketing page. Vendors describe their data residency options in marketing terms that are often more reassuring than the contract language. The data processing agreement is the document that creates a legal commitment. Our guide on what to check in AI vendor contracts covers the specific clauses to look for before signing.

The third step is to check whether your specific plan tier includes the data residency option. Several of the platforms above restrict Australian data centre access to business or enterprise tiers. Using a starter or free tier from the same vendor does not provide the same data residency commitment. Confirm this at the subscription level you are actually on, or plan to be on. For more Australian-specific AI buying guides, reviews, and compliance content across every category, browse our Need to Know AI: Australia hub.

Related reading: OAIC guidance on AI for businesses, HR AI compliance in Australia, free AI staff policy template, AI data breaches and the NDB scheme, Claude AI review for Australian business, Notion AI review for Australian business, and Microsoft Copilot pricing in Australia.

Try our free AI Privacy Risk Scorer to score your current AI tool setup against Privacy Act requirements.

For the questions to ask any vendor about data location, see our data sovereignty hub.

Which AI tools offer Australian data centres?

The main platforms with confirmed Australian data centre options are Microsoft (Azure and Microsoft 365), Google (Google Cloud and Google Workspace), and Amazon Web Services. These cover productivity tools, email, document storage, communication, and cloud infrastructure. For AI model access specifically, AWS Bedrock in the Sydney or Melbourne region gives access to Claude and other models within Australian infrastructure. Most standalone AI tools, including ChatGPT, do not offer Australian data residency at any tier.

Do I need Australian data residency to comply with the Privacy Act?

Not necessarily, but you do need to address your obligations under APP 8 if personal information is disclosed to an overseas recipient. Australian data residency is one way to avoid cross-border disclosure obligations. Another way is to ensure the overseas recipient is bound by enforceable privacy protections that are substantially similar to Australian standards. For most small businesses, the practical approach is either to use a tool with Australian data residency or to get legal advice on whether your current vendor's terms satisfy the APP 8 requirements for your specific use case. See our guide on AI and the Privacy Act in Australia for more detail.

Does Microsoft Copilot store data in Australia?

Yes, for Microsoft 365 tenants provisioned to the Australian region, Copilot operates within your existing tenant environment and inherits its data residency. Core Microsoft 365 data, including content generated by Copilot that is stored in your tenant, sits in the Australian region. A contractual guarantee rather than a default placement requires the Data Residency Advanced add-on. AI model inference routing is covered separately in Microsoft's enterprise agreements and worth confirming if you are in a regulated industry.

Is Google Workspace data stored in Australia?

It can be, with the right plan and configuration. Google Workspace Business Standard and above include a Data Regions policy that lets administrators pin covered data types to Australia. Not all data types are covered by this policy. Google publishes a list of what is and is not included, and it is worth reviewing that list if you are considering this for regulated data. Gemini AI features in Google Workspace operate within your Workspace environment subject to Workspace data processing terms.

Can I use Claude AI and keep my data in Australia?

Yes, if you access Claude through AWS Bedrock in the Sydney or Melbourne AWS region. Data processed through AWS Bedrock in an Australian region stays within that regional infrastructure. Accessing Claude directly at claude.ai or through Anthropic's own API does not currently offer Australian data residency. For businesses with a specific requirement to keep AI-processed data in Australia, AWS Bedrock is the path to achieve that with Claude.

What is the difference between data at rest and data in transit for AI tools?

Data at rest is information stored on the vendor's servers, for example documents in OneDrive or emails in Exchange. Data in transit is information moving between your device and those servers, or between services within the vendor's infrastructure. Most Australian data centre commitments cover data at rest. AI inference, the processing that happens when the AI reads your prompt and generates a response, may occur on different infrastructure and is not always covered by the same data residency commitment. If your obligations require data to stay in Australia at every stage, confirm this explicitly with the vendor rather than assuming data at rest coverage extends to inference processing.