Not All AI Providers Are Equal. Ask These Ten

Here is a sentence that tells you nothing: "our platform is powered by AI." It is now true of almost every business communications product on the market, and it is compatible with at least five completely different architectures, five different accountability structures and five very different answers to the only questions that matter. At one end sits a company that operates its own network, its own platform and its own infrastructure, and can tell you where a particular call was processed on a particular Tuesday. At the other sits a product built in a few months on a public model API, where the entire technical estate is somebody else's and the company you are contracting with cannot change a thing about how your data is handled — because it does not control the thing handling it. Both companies can pass a feature comparison. Both can run an impressive demonstration, because the demonstration is of the model, and the model is very good. The difference only surfaces in three situations: when a data residency review asks where your customer conversations were processed, when something goes wrong and you need it fixed rather than escalated, and when a regulator asks you — not your supplier — to account for an automated decision. This article sets out the five kinds of company, the layers your call audio actually passes through, and ten questions that separate them in about fifteen minutes.

AI Procurement · Infrastructure · 2026

Two Companies, One Claim, Completely Different Risk

A company that operates data centres and a company that operates an API key can write the same sentence on a website. Both statements can be entirely true. The difference between them does not appear in a demonstration, a feature comparison or a price — it appears in a data residency review, in a privacy incident, and at three in the morning when something is broken and you need somebody who can actually change it.

📅 ⏱ 16 min read 🇦🇺 Australian owned, Australian hosted, Australian supported
TL;DR

"We have AI" is compatible with five very different companies. An operator that runs its own network and platform; a platform built on hyperscaler infrastructure; a reseller with a logo on somebody else's product; a thin application over a public model API; and a specialist model provider. All five can truthfully make the same claim, and a demonstration cannot tell them apart because the demonstration is of the model. Your call audio passes through layers — application, middleware, model API, cloud infrastructure — and each layer is a separate company with separate policies. The model provider's protections attach to whoever holds that contract, not automatically to you. Zero data retention announcements in August 2026 are genuine and welcome; they do not tell you what the intermediary you are contracting with does with your audio before or after. Ask ten questions: where processing happens, what is retained and for how long, whether anything trains a model, who the subprocessors are, what happens when the AI is wrong, whether a human can override, who is accountable at 3am, how you get your data out, what your automated-decision disclosure looks like, and what the product does not protect you from. Ask them separately from your general data residency question, because they usually have different answers.

The Claim That Tells You Nothing

Read these four statements. All four are the kind of thing you will find on a communications provider's website in 2026. All four can be literally true.

The claimWhat it is compatible with
"AI-powered transcription and summaries"Anything from a purpose-built local pipeline to a single API call to a service on another continent.
"Enterprise-grade AI security"Usually a restatement of the model provider's security posture, which is not the same as the posture of the company you are paying.
"Your data is never used to train models"Frequently true of the model provider's API. Says nothing about what the intermediary retains, logs, caches or stores on the way through.
"Australian hosted"Accurate about the platform, and silent about the AI feature you enabled with one click, which may be processed somewhere else entirely.

None of these are lies, and that is the difficulty. They are accurate statements at a level of abstraction that hides the thing you need to know. A vendor answering "is your AI secure?" with the security documentation of the model they call is not being deceptive — they are answering the question they were asked, using the strongest evidence they have. The fix is not to distrust vendors. It is to ask a better question, and the rest of this article is about which one.

The Layers Your Call Audio Passes Through

When a call gets transcribed or answered by an AI agent, the audio does not go to "the AI". It travels through a stack, and every layer in that stack is a separate company with separate policies, a separate jurisdiction and a separate security posture.

LayerWhat it isWhose policies govern it
1. The applicationThe product you log into. Handles the call, holds the recording, presents the transcript.The company you contracted with.
2. The middlewareThe glue — queues, storage, logging, retries, caches. The least visible layer and frequently where copies of your audio actually sit.Also the company you contracted with, though they may not have documented it.
3. The model APIThe service that turns audio into text, or generates a response.The model provider, under a contract they hold — not you.
4. The cloud infrastructureThe compute and storage everything above runs on, in a particular country.A hyperscaler or a data centre operator, several contracts away from you.
Layer two is where the surprises live

Most procurement attention goes to layer three, because that is where the recognisable brand names are and where the published policies are strongest. But layer two is where a system keeps queues, retry buffers, error logs containing request payloads, caches and temporary storage — and layer two belongs to the company you are paying, is rarely documented, and is frequently the only place your raw audio is persisted at all. The question "does the model provider retain my audio?" can be answered no while the honest answer to "does my audio sit anywhere for more than a moment?" is yes, in a bucket, for thirty days, in a region nobody has thought about.

Five Kinds of Company That Say "We Have AI"

Each of these is a legitimate business. Each fails a customer in a specific way. Knowing which one you are talking to is most of the work.

🏗️

1. The operator

Runs its own network and platform, and holds direct relationships for each component. Strength: can answer specific questions about specific events, and can change things. Where it fails you: may still call an external model for some AI functions, and must say so plainly rather than letting "we own our infrastructure" imply otherwise.

☁️

2. The hyperscaler-hosted platform

Builds and operates its own software on somebody else's compute. Strength: genuine engineering, real control over its own code, good scalability. Where it fails you: region selection is a configuration choice, and an outage of the underlying provider is an outage they can only wait out with you.

🏷️

3. The reseller

Sells another company's platform under its own name. Strength: often excellent local sales and account management, and there is nothing wrong with buying this knowingly. Where it fails you: cannot change the product, cannot fix a fault, and cannot tell you what happens at layers two to four because they do not operate any of them.

🪶

4. The thin application layer

A product built quickly over a public model API. Strength: genuinely fast innovation, and often the nicest interface in the market. Where it fails you: the entire estate is somebody else's, the whole business depends on one API's pricing and terms, and a policy change upstream arrives without notice.

🧠

5. The model provider

Builds and runs the models. Strength: the strongest published security and retention commitments in the chain, by a wide margin. Where it fails you: you are almost certainly not their customer. Their commitments run to whoever holds the contract, and that is your supplier, not you.

🔍

How to tell them apart

One question does most of it: "if I report a fault at 3am that turns out to be in the AI transcription, who fixes it and what is their name?" The answer separates these five faster than any amount of documentation.

What You Inherit, and What You Do Not

This is the conceptual error at the centre of most AI procurement, and it is worth spelling out because it is easy to make and expensive.

What people assumeWhat is actually the case
"The model provider does not train on API data, so my data is safe."That commitment runs to the entity holding the API contract. It says nothing about what your supplier does with your audio before it is sent or after it comes back.
"The model provider offers zero data retention, so nothing is retained."Nothing is retained at that layer, if the option is enabled by the contract holder. Layers one and two are unaffected and belong to your supplier.
"Their compliance certifications cover us."Certifications describe the certified party's controls. They do not transit down a supply chain, and your obligations do not transit up it.
"They are a big company, so it is fine."The size of a company four contracts away is not a control. The relevant question is what the company you can actually telephone controls.

You inherit the security posture of every layer, and the accountability of exactly one. That asymmetry is the whole reason this matters: the risks are distributed across four companies and the remedy is available from one. Choosing a supplier is therefore mostly a question of how much of the stack that one company can actually reach.

The August 2026 Developments, and Their Actual Scope

The model providers have moved genuinely and recently, and it deserves credit before it gets qualified.

In August 2026, zero data retention arrangements became available for frontier models, meaning prompts and responses are not retained after a request is processed. Alongside that, customer-controlled deployments keep content on customer-managed infrastructure, and work is under way on a mode where content is stored on the provider's infrastructure but encrypted with customer-held keys, so provider personnel cannot read it. Enterprise API data is not used for model training unless a customer explicitly opts in.

This is real progress. Here is its scope.

All of the above is a substantial improvement and businesses should ask for it. What it does not do is tell you anything about the layer you are contracting with. Zero data retention is an option somebody must enable, in a contract you are not party to, at a layer that is one of four. The published analysis of enterprise AI risk makes exactly this point: data flows through multiple layers — application, middleware, model API, cloud infrastructure — and vulnerabilities appear in that chain even where encryption is used, because the weakest architecture in the chain sets the real posture regardless of how strong the model provider's policies are.

So the correct use of these developments in procurement is not "our AI is safe because the model provider announced ZDR". It is: "is zero data retention enabled on the contract that carries my audio, and what does your own middleware do with it either side of that call?" A supplier who can answer both halves is operating the stack. A supplier who can only answer the first half is quoting somebody else's homework.

The Question That Gets Asked Wrong

Nearly every Australian business evaluating a phone system asks about data residency. Nearly all of them ask it in a way that produces a true and useless answer.

The question usually askedWhy the answer misleads
"Is my data hosted in Australia?"Answers for the platform. The AI feature is frequently a separate path with a separate answer, and the responder is thinking about the platform.
"Are you an Australian company?"Corporate domicile has almost no relationship to where processing occurs.
"Do you comply with Australian privacy law?"Everybody says yes, and the answer is compatible with lawful offshore disclosure under APP 8.
Ask this instead, and ask it separately

"When a call is transcribed, summarised or answered by an AI agent — where does the audio or text go, which company processes it, is it retained by that company, for how long, and is it used to train models?"

Ask it as a distinct question from general data residency, because they usually have different answers. A provider who has thought about this answers immediately and specifically. A provider who has not will offer to find out — which is a perfectly honest response and is also, itself, the answer to your real question.

The Ten Questions

Fifteen minutes, no technical knowledge required, and they work on us as well as on anybody else.

#QuestionWhat a strong answer looks like
1Where is the audio processed, physically, for each AI feature?A country per feature. Different answers for transcription and for an AI agent is fine and honest; a single vague answer for all of it is not.
2What is retained, at which layer, and for how long?Separate answers for the platform, your middleware and the model API. Specific periods, configurable, in writing.
3Is anything used to train a model, ever?No by default, stated in the contract rather than in a marketing page or a blog post.
4Who are your subprocessors for AI functions?A list, maintained, with notification of changes. If they will not name them, you cannot assess them.
5What happens when the AI is wrong?A defined failure path — escalate to a human, log the event, notify somebody. Not an assumption of correctness.
6Can a human always override, and is there always a path to a person?Yes, unconditionally. Anything less is disqualifying for anything consequential.
7If I report a fault at 3am in the AI layer, who fixes it?A named team with the ability to change the thing that is broken. "We raise it with our vendor" is an answer, and it tells you which of the five companies you are talking to.
8How do I get my data out if I leave?An export format, a process, a timeframe. Recordings and transcripts included, not just contacts.
9What does your automated-decision disclosure look like?Evidence they know about the 10 December 2026 obligation. If they have not thought about it, they will not help you meet it.
10What does this not do, and where does it fail?A real answer. A supplier who cannot name their product's limits has not examined it, and this question is the single best predictor of the rest.

The Evidence to Ask For

Answers are cheap. These four artefacts are not, and asking for them changes the conversation immediately.

📄

A data flow description

One page, plain language, showing where audio goes for each AI feature and which company handles it at each step. Any supplier operating their own stack can produce this in a day.

📋

A current subprocessor list

Names, functions, jurisdictions, and a commitment to notify you of changes. Unnamed subprocessors are unassessable subprocessors.

The retention schedule

Per data type — recordings, transcripts, summaries, logs, metadata — with the configurable periods marked. Logs are the category everybody forgets and they frequently contain content.

📞

The incident and escalation path

Who you call, in what hours, in which country, and what they can change. Ask for the after-hours version specifically, because the business-hours version always looks better.

Six Answers That Should Worry You

If you hear thisWhat it usually means
"We use enterprise-grade AI with bank-level security."A description of somebody else's product, offered because the speaker does not know the details of their own.
"Our AI is proprietary, so we cannot discuss the architecture."Sometimes genuine commercial protection. Frequently an inability to answer. You can distinguish these by asking a narrower question and seeing whether it gets narrower too.
"All data stays in Australia" — with no qualification about AI features.Either a strong claim they can substantiate in a day, or a claim nobody has tested against the AI path. Ask for it in writing and you will find out which.
"We detect deepfakes and synthetic voice reliably."Overconfidence about a capability that is not currently reliable. It calls the rest of their claims into question.
"That has never happened."An answer about the past presented as a control for the future. Ask what the process is when it does.
"We'll come back to you on that." — for every one of the tenNot disqualifying on one question. Disqualifying on all ten, because it means nobody has ever asked and nobody has ever thought about it.

The Three in the Morning Test

Strip away the documentation and this is what you are actually buying.

Something is wrong. AI transcripts are silently truncating, or the AI agent is misrouting urgent calls, or a summary contains a customer's details from a different account. You call your provider. Three things then determine what your week looks like, and none of them appears in a feature comparison.

  1. Can the person who answers see the thing that is broken? Not a status page — the actual system, with your call in it. An operator can. A reseller cannot, and a thin layer can see only its own half.
  2. Can anybody in that company change it? If the fault is upstream, your provider's role is to file a ticket and wait alongside you. That is not their failing; it is the architecture you bought.
  3. Is anybody in your timezone empowered to make a decision? Turning a feature off, failing back to a safe mode, or accepting a workaround are decisions, and they need somebody awake and authorised.

This is the whole argument for infrastructure, and it is an accountability argument rather than a technical one. Owning infrastructure does not make software better. What it does is remove the gap between two companies where a fault can sit unresolved while each correctly certifies its own half. The fewer contracts between you and the thing that is broken, the shorter that gap — and at three in the morning, gap length is the only specification that matters.

The Obligation That Stays With You

One asymmetry deserves stating flatly, because a lot of businesses discover it at the worst moment.

Your obligations do not transfer to your supplier

Recordings and transcripts of your calls are personal information, and where they are processed is part of your compliance position, not merely part of your provider's architecture diagram. Cross-border disclosure obligations attach to you. From 10 December 2026, if you use personal information in automated decision-making capable of affecting a person's rights or interests, your privacy policy must disclose the kinds of information used and the kinds of decisions made — and an AI agent that qualifies leads, prioritises a queue or decides who reaches a human quickly is plausibly within scope. Your supplier can help you answer these. They cannot answer them for you, and no contract makes them the regulator's counterparty instead of you.

Which is why question nine on the list is not a formality. A supplier who has never considered automated-decision disclosure will not be able to tell you what to put in your privacy policy, and December is closer than it looks.

In Fairness to the Thin Layer

It would be easy to end with "buy from an operator", and that is not quite the honest conclusion.

The fastest genuine innovation in this market is coming from small teams building over public model APIs. They ship faster than anybody else, they are frequently the nicest products to use, and for a low-stakes internal use case with no sensitive content, the architecture concerns in this article are largely academic.

A thin layer is a reasonable choice when...It is the wrong choice when...
The use case is internal and low-stakesCustomer conversations are being processed
No sensitive or regulated content is involvedYou hold health, financial, legal or personal information
You can absorb a change of terms or an outageThe function is on the critical path for revenue or safety
You are experimenting, deliberately, with an exit in mindYou will be asked to demonstrate what happened, to whom, and where

The failure is not choosing a thin layer. It is choosing one while believing you chose an operator — and that mistake is made in procurement, not in engineering, because both companies wrote the same sentence on their website.

Where We Stand, and What We Will Put in Writing

It would be poor form to publish ten questions and dodge them, so here is our position.

We are Australian owned, Australian hosted and Australian supported, and we operate our own platform rather than reselling somebody else's. For AI features specifically, we will tell you — per feature, in writing — where the processing happens, which company performs it, what is retained and for how long, and whether anything is used for training. Where a component is not ours, we will say so plainly rather than letting "we own our infrastructure" imply something broader than it should. That distinction is exactly the one this article is about, and we would be poorly placed to fudge it.

Ask us the ten questions, then ask everyone else

Send them to us in an email and we will answer them in an email, per feature, specifically. Then send the same email to whoever else you are considering and compare the two replies rather than the two brochures.

Get Started Or call 1300 881 662
The summary

Five very different companies can truthfully say they have AI, and a demonstration cannot distinguish them because the demonstration is of the model. Your call audio traverses four layers owned by different companies, and the middle layer — your supplier's own middleware, with its queues, caches and logs — is where the surprises usually are. The model provider's protections run to whoever holds that contract, which is not you. August 2026's zero-data-retention developments are real and are not an answer about your supplier. Ask the AI processing question separately from general data residency, because they usually have different answers. And ask who fixes it at 3am, because that single question sorts the five categories faster than anything else — and because the obligations stay with you regardless of the answer.

Related reading: Australian infrastructure versus overseas cloud, which covers the general data residency case and the AI exception, the December 2026 automated-decisions obligation, how transcription and summaries actually work, where AI genuinely improves your security, and what AI voice agents cost and return.

Frequently Asked Questions

What does it mean when a phone system provider says they have AI?
Almost nothing on its own, because the claim is compatible with at least five very different kinds of company. First, an operator that runs its own network and platform and holds direct relationships for each component, which can answer specific questions about specific events and can change things when they break. Second, a platform that builds and operates its own software on somebody else's cloud compute, which involves genuine engineering and real control over its own code but where region selection is a configuration choice and an underlying provider outage is something they can only wait out alongside you. Third, a reseller selling another company's platform under its own name, which often comes with excellent local account management but cannot change the product, cannot fix a fault, and cannot describe what happens deeper in the stack because it operates none of it. Fourth, a thin application layer built quickly over a public model API, which innovates fastest and frequently has the nicest interface, but where the entire technical estate belongs to someone else and an upstream terms change arrives without notice. Fifth, the model provider itself, which has the strongest published commitments in the chain — and whose customer is almost certainly your supplier rather than you. One question sorts them: if a fault at 3am turns out to be in the AI layer, who fixes it and what is their name?
Where does my call audio actually go when a call is transcribed by AI?
Through a stack of four layers, each owned by a different company with different policies, jurisdictions and security postures. Layer one is the application you log into, which handles the call, holds the recording and presents the transcript, governed by the company you contracted with. Layer two is the middleware — the queues, storage, logging, retries and caches that glue everything together — which also belongs to your supplier but is frequently undocumented. Layer three is the model API that turns audio into text or generates a response, governed by the model provider under a contract they hold and you do not. Layer four is the cloud infrastructure everything runs on, in a particular country, several contracts away from you. Most procurement attention goes to layer three because that is where the recognisable brands and the strongest published policies are, but layer two is where the surprises live: retry buffers, error logs that contain request payloads, caches and temporary storage. It is entirely possible for the answer to does the model provider retain my audio to be no, while the honest answer to does my audio sit anywhere for more than a moment is yes, in a bucket, for thirty days, in a region nobody has considered. Ask about each layer separately.
If the AI provider says they do not train on my data, is my data safe?
Not necessarily, because that commitment attaches to whoever holds the contract with the model provider, and that is your supplier rather than you. Enterprise API terms commonly do state that customer data is not used for model training unless the customer explicitly opts in, and that is genuine and worth having. What it does not describe is what your supplier's own application and middleware do with your audio before it is sent upstream and after it comes back — the logging, caching, queueing and storage at layers one and two, which are entirely within your supplier's control and outside the model provider's terms. The same reasoning applies to certifications: a certification describes the certified party's controls, it does not transit down a supply chain, and your own obligations do not transit up it. Similarly, the size or reputation of a company four contracts away from you is not itself a control. The practical framing is that you inherit the security posture of every layer in the chain and the accountability of exactly one company, and the weakest architecture anywhere in that chain sets your real risk regardless of how strong the model provider's policies are. So ask your supplier the retention question about their own systems specifically, and ask for the answer in writing.
What is zero data retention and does it solve the problem?
Zero data retention means the model provider does not retain customer prompts or model responses after a request has been processed, and arrangements of this kind became available for frontier models in August 2026, alongside customer-controlled deployments that keep content on customer-managed infrastructure and work on a mode where content is stored on the provider's infrastructure but encrypted with customer-held keys so provider personnel cannot read it. This is real progress and businesses should ask for it. It does not solve the problem on its own for two reasons. First, it is an option that somebody must actually enable, in a contract you are not a party to, at one of four layers. Second, and more importantly, it says nothing at all about the layer you are contracting with — the published analysis of enterprise AI risk makes exactly this point, that data flows through application, middleware, model API and cloud infrastructure, and vulnerabilities appear along that chain even where encryption is used. So the correct use of these developments in procurement is not our AI is safe because the model provider announced zero data retention. It is: is zero data retention enabled on the contract that carries my audio, and what does your own middleware do with it either side of that call? A supplier who can answer both halves operates their stack.
How should I ask about data residency for AI features?
Ask about AI processing as a separate question from general data residency, because they usually have different answers and the person answering is thinking about the platform. The three questions businesses normally ask all produce true and useless answers. Is my data hosted in Australia answers for the platform, while the AI feature is frequently a separate path. Are you an Australian company asks about corporate domicile, which has almost no relationship to where processing occurs. Do you comply with Australian privacy law gets a yes from everybody and is compatible with lawful offshore disclosure. The question that works is longer and more specific: when a call is transcribed, summarised or answered by an AI agent, where does the audio or text go, which company processes it, is it retained by that company, for how long, and is it used to train models? Ask for a country per feature, since different answers for transcription and for an AI agent are honest and expected, whereas one vague answer covering everything is not. A provider who has thought about this will answer immediately and specifically. A provider who has not will offer to find out, which is a perfectly honest response and is also, in itself, the answer to your underlying question about how carefully this has been considered.
Is it ever fine to use a small AI provider built on a public API?
Yes, and pretending otherwise would be dishonest, because the fastest genuine innovation in this market is coming from small teams building over public model APIs. They ship faster than anyone else and frequently produce the nicest products to use. A thin layer is a reasonable choice when the use case is internal and low-stakes, when no sensitive or regulated content is involved, when you could absorb a change of terms or an extended outage without material harm, and when you are experimenting deliberately with an exit already in mind. It is the wrong choice when customer conversations are being processed, when you hold health, financial, legal or personal information, when the function sits on the critical path for revenue or safety, or when you may be asked to demonstrate what happened, to whom and where. The failure mode is not choosing a thin layer — it is choosing one while believing you chose an operator, and that mistake happens in procurement rather than in engineering, because both kinds of company can write exactly the same sentence on their website. So decide deliberately which category you are buying from, and if it is a thin layer, keep the scope matched to that decision and keep the exit path real.
What AI obligations stay with my business rather than my provider?
More than most businesses expect, and no contract shifts them. Recordings and transcripts of your calls are personal information, and where they are processed forms part of your compliance position rather than merely part of your provider's architecture. Cross-border disclosure obligations attach to you as the entity that collected the information, not to the supplier who moved it. From 10 December 2026, if you use personal information in automated decision-making capable of affecting a person's rights or interests, your privacy policy must disclose the kinds of personal information used and the kinds of decisions made — and an AI agent that qualifies leads, prioritises a queue, decides who reaches a human quickly, or scores an interaction is plausibly within scope, as the drafting is broad enough to capture rule-based tools and automated assessment technologies as well. Your supplier can help you answer these questions and a good one will. They cannot answer them for you and they do not become the regulator's counterparty in your place. This is why asking a prospective supplier what their automated-decision disclosure position looks like is not a formality: a supplier who has never considered it will be no help in drafting yours, and the date is closer than it appears once you account for needing an inventory of every automated decision your systems already make.

What to Read Next

Your next reads

Uniden Voice Over Cloud logo

Australia’s smartest AI-powered cloud phone system — Australian owned, Australian hosted, Australian supported. unidenvoice.com | 1300 881 662