AI Agents for Business: The 2026 Deployment Guide

There is a version of this that goes badly and it usually starts with enthusiasm. Somebody sees a demo where an AI answers a call, understands a slightly awkward sentence, looks something up, books an appointment and writes a note, and it is genuinely impressive, because that demo is real and the technology behind it works. So the business buys it, points it at the main number, and switches it on for everything. Three weeks later it is off again. Not because it could not understand people, but because nobody decided what it was allowed to promise, nobody wrote down what should happen when it was unsure, nobody told the team it was coming, and nobody agreed in advance what a good week would look like, so when a customer complained about one call there was no way to weigh that against the ninety it handled properly. That is the failure pattern behind Gartner's estimate that over forty per cent of agentic AI projects will be cancelled by the end of 2027, and the three causes it names are escalating cost, unclear business value and inadequate risk controls. Notice what is not on the list. The models are not the problem. The scoping is. An AI agent is not a feature you enable, it is a job you hand over, and jobs get handed over properly or they get handed back. This is how to hand one over properly: how to choose the first one, what to build around it, how to release it over ninety days without betting the main number on week one, and which five numbers tell you whether it is working.

Implementation Guide · Agentic AI 2026

An AI Agent Is a Job Not a Feature You Switch On

Gartner expects more than four in ten agentic AI projects to be cancelled by the end of 2027, and the reasons given are not technical: escalating cost, unclear business value, and risk controls that were never built. Meanwhile the average Australian organisation using AI is running eleven agents at once, and voice AI is handling close to a fifth of inbound contact centre volume. Both things are true, and the difference between them is almost entirely in how the first agent was scoped. This is the scoping, the guardrails, the ninety day sequence and the numbers you measure afterwards.

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

An agent is different from a chatbot and from an automation. A chatbot answers. An automation follows a fixed path. An agent is given a goal, decides the steps, uses tools to get there, and knows when to stop and hand over. Pick a first job with high volume, low variance and low consequence. After hours answering, booking, order and job status, and writing call notes back into the CRM are where most Australian businesses get a payback inside a quarter. Build six things, not one: the job definition, the tools it may use, the context it can see, the guardrails it cannot cross, the escalation path, and the log. Most failed projects built the first and skipped the other five. Release it over ninety days: shadow first, then after hours only, then supervised daytime overflow, then the front line. Measure five numbers and one of them is repeat contact, because containment that just pushes the caller into ringing back tomorrow is not containment. Beware agent washing. Gartner's view is that only a small fraction of vendors claiming agentic capability actually have it, and the tell is what happens when the caller says something the script did not anticipate.

What an Agent Is, and What It Is Not

The word has been stretched far enough that it is worth being blunt about the distinction, because the three things get sold under the same name and they behave very differently when a customer says something unexpected.

ThingWhat it doesWhat happens when the caller goes off script
A chatbot or IVRMatches what it hears against a list of expected inputs and returns a prepared response or route.It fails, loops, or sends the caller to a default. It cannot recover, because there was never anything behind the list.
An automationRuns a fixed sequence when a trigger fires. Missed call sends an SMS. Form submission creates a task.Nothing. It runs the same sequence regardless, which is fine when the trigger is unambiguous and useless when it is not.
An agentHolds a goal, decides which steps to take, calls tools to look things up and change things, and judges when it is done or when it should stop.It reasons about it. It can ask a clarifying question, look something up it was not told to look up, or decide this one belongs with a person and hand it over with the context intact.

The practical test is whether the thing can do something it was not explicitly told to do, in service of a goal it was told to achieve. A caller who says "I rang yesterday about the leak at the Brunswick site and nobody has been out" is giving an agent three facts and an implicit request. A menu hears none of it. An automation was not triggered by it. An agent can look up the job, see the technician was booked for this morning, check whether that appointment was completed, and either tell the customer where the technician is or escalate to the person who can. That is the whole difference, and it is why the term caught on.

Autonomy is a dial, not a switch

The useful question is never "is it autonomous". It is "which decisions is it allowed to make on its own, and which ones does it have to bring to somebody". A well-scoped agent might be fully autonomous about looking up a booking, allowed to change one with confirmation, and forbidden from issuing a credit at all. Three different levels inside one job. Every agent that has caused a business a real problem was given a single level of autonomy for the whole job, usually the highest one.

Where This Actually Sits in 2026

The gap between the headline numbers and the deployment numbers is the single most useful thing to understand before you start, because it tells you which risk you are actually running.

What the numbers sayWhat it means for you
Gartner expects over 40% of agentic AI projects to be cancelled by the end of 2027, citing escalating cost, unclear business value and inadequate risk controls.The failure mode is commercial and operational, not technical. All three causes are fixed at scoping time, before anybody signs anything.
Around 17% of organisations have deployed agents, while over 60% expect to within two years.You are not late. You are in the window where being deliberate still beats being first, and where the people who rushed are currently rebuilding.
Salesforce puts the average Australian organisation using AI at around eleven agents running at once.The organisations getting value are running many small agents with narrow jobs, not one large one with a broad one. That is the shape to copy.
Voice AI is reported to handle roughly 19% of inbound contact centre volume in 2026, against about 6% in 2024.Voice moved from experiment to production in about eighteen months. Customer tolerance for it has moved with it, provided it works.
A 2026 Gartner survey found 91% of service leaders under executive pressure to implement AI, and while around 88% of contact centres use AI in some form, only about a quarter have integrated it properly.Most deployments are bolted on beside the systems rather than joined to them. That gap is where the disappointment comes from, and it is an integration problem.
Gartner's longer view has agentic AI autonomously resolving around 80% of common service issues by 2029, with roughly 30% lower operating cost.A direction of travel, not a target for your first quarter. Businesses that aim at 80% in year one are the ones writing the cancellation notices in year two.

Read those together and the conclusion is uncomfortable for the enthusiasts and reassuring for everybody else. The technology is ready enough that real volume is flowing through it. The implementations are not, and the difference between the two groups is almost entirely process. Nobody is cancelling these projects because the model could not understand an Australian accent.

The Six Parts of a Working Agent

When a deployment fails, it is nearly always because one to five of these were never built. The first one is easy and everybody does it, which is why it feels like the whole job.

1

The job

Written in one paragraph, in plain words, including what success looks like and what is out of scope. If it takes a page, the job is too big. "Answer calls after 5pm, identify the caller, answer questions about opening hours, services and job status, book an appointment if asked, and take a message for anything else."

2

The tools

The specific things it is allowed to call: look up a customer, read the calendar, create a booking, create a task, send an SMS. Each one named, each one with a defined input and output. An agent with no tools is a chatbot with good manners.

3

The context

What it can see when the call starts. The caller's number matched to a customer record, their last three interactions, their open jobs, their account status. Context is what makes it feel like the business rather than like a stranger.

4

The guardrails

The things it may never do, stated as prohibitions rather than as guidance. No pricing outside the published list. No commitments about timeframes. No account changes without verification. No discussing another customer.

5

The escalation

Where it goes when it should stop, what it hands over, and what the customer hears while that happens. This is the part that decides whether people trust it, and it gets about a tenth of the attention it deserves.

6

The log

Every call, every tool it called, every decision point, every escalation, kept where you can read it. Without this you cannot improve it, cannot defend it, and cannot answer the first complaint, which will arrive in week two.

If you only take one thing from this article: the six parts are not optional extras that mature deployments add later. A project that builds part one and skips parts four, five and six is not an early version of a good deployment. It is the thing Gartner is counting when it says forty per cent get cancelled. Guardrails and escalation are cheaper to build before launch than after the first bad call, and the log is the only reason you will ever know which calls were bad.

Choosing the First Job

Three variables decide whether a job is a good first candidate, and you want the same answer on all three. High volume, low variance, low consequence.

VariableWhat you wantWhy
VolumeHigh. At least a few dozen a week.Low volume jobs cannot repay the setup, and worse, they never generate enough examples for you to tell whether it is working. You need repetition to learn anything.
VarianceLow. The same handful of shapes over and over.High variance work is where agents are impressive in demos and inconsistent in production. Start where the ground is flat and expand into the rough later.
ConsequenceLow. A mistake is recoverable within a day.Not because agents make more mistakes than people, but because your process for catching and fixing mistakes does not exist yet. Build that muscle somewhere forgiving.

Run your own candidate list through those three and you usually find that the job everybody wants to automate first, the complicated one that eats a senior person's week, is exactly the wrong one to start with: low volume, high variance, high consequence. It is also the one that will be automatable in about a year, once you have the plumbing, the guardrails and the institutional habit of supervising an agent. Start with the boring one that happens two hundred times a week.

The success test has to be written before you build

One sentence, with a number in it, agreed by the person who owns the outcome. "By day sixty, at least half of calls arriving after 5pm are fully dealt with without a callback the next morning." If you cannot write that sentence, you do not yet have a project, you have an interest. And if the sentence is written after launch, it will be written to describe whatever happened, which is how projects survive for a year without anybody being able to say whether they worked.

Five Jobs That Pay Back Inside a Quarter

These are the ones we see returning their cost fastest in Australian businesses between roughly five and two hundred staff. They are unglamorous on purpose.

JobWhat the agent doesWhat it replaces
1. After hours answeringAnswers, identifies the caller, deals with the common questions, books or takes details, and flags anything genuinely urgent to the on-call person with context.A mailbox nobody opens until morning, or an unbounded ring that records nothing at all. This is the single largest quick win for most businesses.
2. Booking and reschedulingReads the calendar, offers real times, books, confirms by SMS, and handles the reschedule when the customer rings back.Phone tag. Roughly half of booking calls in most businesses are a reschedule, and reschedules are the most mechanical conversation in the business.
3. Job and order statusMatches the caller, finds their job or order, tells them where it is, and offers to notify them when it changes."Let me find out and call you back", which almost always costs two calls and a note that never gets written.
4. Call notes and CRM updatesSummarises the call, extracts the actions, writes them to the right record, and creates the follow up task.The note that nobody wrote. This one is quietly the highest value of the five, because it fixes data quality everywhere downstream.
5. Overflow at peakTakes the calls that would otherwise queue past a threshold, handles what it can, and offers a callback that keeps the caller's place.Abandoned calls, which are invisible in most reporting and are the most expensive thing a busy business does all day.

Notice that four of the five are about calls that are currently not being handled at all rather than calls currently handled by a person. That is deliberate, and it is the shortcut past the hardest internal conversation. An agent that picks up work nobody is doing does not need anybody to give anything up, which means it gets a fair trial. An agent introduced as a replacement for a person's work gets audited by that person, and they will find the three calls it got wrong before they find the ninety it got right.

Guardrails: What It May Never Do

Write these as flat prohibitions, in a list, with no conditions attached. Guidance gets interpreted. Prohibitions get enforced.

GuardrailWhy it matters
No pricing beyond the published listA quoted price on a recorded call is a representation. Under Australian Consumer Law it is yours, whoever said it. The agent may read the published rate card and must refuse to estimate anything else.
No commitments about time"Someone will be there this afternoon" is the most common promise an over-eager agent makes and the one that generates complaints. It may state a booked appointment. It may not predict one.
No account changes without verificationDefine the verification, define what a failed verification does, and make sure the failure path is not simply a second attempt with an easier question.
No discussion of any other customer or jobAgents with broad data access can be talked sideways into adjacent records. Constrain the tool, not just the instruction: the lookup should only return records tied to the identified caller.
No claim to be humanIt should not lie if asked, and it is better that it says what it is at the start. Businesses that disclose it get fewer complaints than businesses whose customers work it out on their own.
No handling of a caller in distressDefine the trigger words and route straight to a person. This is not a place for containment metrics, and no business has ever regretted escalating one of these too early.
No unbounded retriesTwo attempts to understand, then escalate. A third loop is how an agent turns a mild frustration into a complaint about the business.
Enforce guardrails in the tools, not only the instructions

An instruction that says "do not issue refunds" is a preference. Not giving the agent a refund tool is a control. Where the consequence is real, remove the capability rather than asking for restraint, and reserve instruction-level guardrails for things you cannot express structurally. This one architectural habit prevents most of the incidents that make the news.

Escalation Is the Hard Part

Everybody builds the happy path. Escalation is where deployments are actually judged, because the calls that get escalated are disproportionately the calls that matter, and a bad handover is worse than no agent at all.

It needs four triggers, at minimum. The customer asks for a person, and the answer is always yes, immediately, without a negotiation. The agent is not confident, on its own assessment. The caller has repeated themselves, which is a reliable signal of a conversation going badly. And the topic is on the never list, such as a complaint, a dispute, or distress.

It needs to carry the context across. The person who picks up should see who the caller is, what they have said so far, what the agent did, and why it escalated. An escalation that makes the customer start again is the moment they decide the whole thing was a waste of their time, and they are not wrong.

It needs a defined destination at every hour of the day. Escalating at 2am to a ring group that is empty is not an escalation, it is a hang up with extra steps. If there is genuinely nobody, the honest version is a commitment with a time attached and an SMS confirming it.

And the customer needs to hear what is happening. "I will put you through to someone who can sort that out, it will take about thirty seconds" costs nothing and changes the entire experience of being handed over.

The Ninety Day Sequence

Four phases, each with an exit test. The point of the sequence is that you are never betting more than you can recover from in a day, and that by the time it is on the front line it has already handled thousands of real conversations under supervision.

PhaseWhat happensExit test
Days 1 to 14: shadowThe agent runs against real calls and produces what it would have said and done, but nothing reaches a customer and no tool it calls changes anything. A person reads a sample every day.You have read at least a hundred transcripts and you can predict what it will do. Every surprise has become either a guardrail or a fix.
Days 15 to 35: after hours onlyLive, but only on calls that would currently reach a mailbox. The downside is bounded because the alternative was nothing.Escalation works at 2am. Notes are landing in the right records. Nobody in the team has found a call that embarrassed the business.
Days 36 to 60: supervised overflowDaytime, but only calls that would otherwise queue past a threshold. Somebody owns the review, half an hour a day, and the team can flag a call in one click.Containment is stable across three weeks and repeat contact is flat or down. Not rising containment. Stable, with repeat contact behaving.
Days 61 to 90: front line, narrowIt answers first on the defined job and hands over everything else. Autonomy widens one decision at a time, each with its own review.The success sentence you wrote in week one is either met or honestly not met. Both are useful. Only ambiguity is a failure.

Three things run alongside all four phases. The team is told before it goes live, what it does, and how to flag a bad call, because staff who discover an agent by accident become its most motivated critics. A named person owns the daily review, and it is in their workload rather than on top of it. And every change is logged with a date, so when something shifts in week nine you can find out what you changed in week eight.

Five Numbers, and Why One of Them Is Repeat Contact

Containment on its own is the number that makes projects look successful right up until somebody checks the callback volume.

NumberDefinitionWhat a healthy answer looks like
ContainmentShare of conversations finished by the agent without a person.Rising, then settling. A first job in the right range lands somewhere between 40% and 70%. Anything near 100% means the scope is trivially narrow or the escalation is broken.
ResolutionShare of contained conversations where the customer's actual purpose was achieved.This is the honest sibling of containment and it needs sampling, not just logs. Read fifty calls a month and score them.
Repeat contact in 72 hoursShare of contained conversations where the same customer makes contact again about the same thing.Flat or falling. If this rises while containment rises, the agent is not resolving anything, it is deferring. This is the single most important number here.
Escalation qualityOf escalated calls, the share where the person who took over had the context and did not have to start again.Above 90%, and it is a configuration problem rather than an AI problem if it is not.
Cost per resolved contactAll-in cost divided by resolved contacts, including the review time.Compared against your loaded cost of a staffed contact. Include the supervision, or the comparison is dishonest and you will be caught out at the first budget review.

Containment that pushes work into tomorrow is not containment, it is a queue with a delay. A caller who was told something unhelpful at 7pm and rings back at 9am has been counted as a success by most vendor dashboards and has cost you two contacts instead of one, plus the goodwill. Pair every containment figure with repeat contact for the same period, always, and treat a divergence between the two as the first thing to investigate rather than the last.

Connecting It to Everything Else

The finding that 88% of contact centres use AI while only about a quarter have integrated it properly is the whole story of wasted spend in this category. An agent that cannot see your systems is doing an impression of your business from the outside.

What it needs to read. Customer records, open jobs or orders, the calendar, the knowledge the team uses, and the history of previous contact. Most of the perceived intelligence of a good agent is actually context. An agent that knows the caller has an open job and a technician booked for tomorrow sounds vastly cleverer than one that does not, using the same model.

What it needs to write. Notes on the record, tasks for people, bookings in the calendar, and updates to job status. Write access is where the value lives, because reading saves the customer time and writing saves your team's.

How it should connect. Through documented APIs and webhooks in both directions, so the agent is not locked to one vendor's marketplace. The Model Context Protocol has become the common way for agents to reach tools and data since it emerged, and it is now under independent stewardship with the major AI ecosystems supporting it. You do not need to care about the specification. You do need to ask a vendor whether their platform can expose your systems to an agent through an open standard, or whether every connection is a bespoke build they charge for.

This is also the argument for putting the communications platform at the centre rather than at the edge. The conversation is where intent shows up first, and a platform that already handles the call, the SMS and the messaging channels is the natural place for the agent to sit, with the systems of record around it. The case for one connected platform covers that in more detail.

The Australian Rules as They Stand

The regulatory position changed direction in the last eighteen months and a lot of advice written in 2024 and 2025 is now wrong in a way that matters commercially.

What people expectWhere it actually stands
Mandatory guardrails for high risk AIProposed in September 2024, then dropped. The National AI Plan of December 2025 confirmed Australia would rely on existing law and sector regulators supported by voluntary guidance, rather than a standalone AI Act.
An Australian AI ActNot in force. In July 2026 the Government set out a direction including legislating Australian Standards for AI and established an Office of AI within the Department of the Prime Minister and Cabinet. Watch it, do not wait for it.
Nothing applies, thenWrong, and this is the expensive misunderstanding. Privacy law, the Australian Consumer Law, record keeping obligations, industry codes and the telecommunications rules all apply to what your agent says and does, exactly as they apply to a staff member.
Government rules are the same as business rulesThey are not. Mandatory requirements including AI impact assessments and Chief AI Officers began applying to Commonwealth agencies from 15 June 2026, with full compliance due 10 December 2026. Useful as a template, not binding on a private business.
Automated decisions are unaffectedFrom 10 December 2026, privacy policies must disclose the kinds of personal information used in substantially automated decisions and the kinds of decisions made. If your agent decides anything about a person, this reaches it.

The practical reading for a business deploying an agent on the phone: your obligations come from the law you already had. Recording consent rules did not change. Consumer law did not change. Privacy did change, in a specific and disclosable way, with a date on it. We covered that in detail in the Privacy Act automated decisions guide, including the part most advisory pieces get wrong, which is that the obligation is disclosure rather than a prohibition or a right to human review.

Eight Ways This Goes Wrong

FailureThe fix
1. The first job was too bigNarrow it until it is almost embarrassing, then launch. The second agent takes a fortnight because the plumbing exists.
2. No success sentenceWrite it, with a number and a date, before you build. Agreed by the person who owns the outcome.
3. Guardrails written as guidanceConvert to prohibitions, and enforce the serious ones by withholding the tool rather than asking nicely.
4. Escalation goes nowhere at nightDefine a destination for every hour. A commitment with a time beats a silent handover to an empty group.
5. Nobody reads the transcriptsHalf an hour a day, named owner, inside their workload. This is where every improvement in the first quarter comes from.
6. It is not connected to anythingGive it read and write access to the customer record and the calendar. Context is most of what makes it work.
7. The team found out when a customer told themTell them first, show them the transcripts, give them a one click way to flag a bad call, and act on the flags visibly.
8. Containment celebrated in isolationPair it with repeat contact every single time it is reported. Make them one line on the report, not two.

How to Tell a Real Agent From a Rebadged Menu

Gartner named the practice agent washing: rebranding assistants, robotic process automation and chatbots as agentic without the underlying capability, and its assessment is that only a small fraction of the vendors making the claim have it. You do not need to evaluate architecture to find out. You need five questions and one phone call.

AskWhat a real answer sounds like
"Which tools can it call, and can I add my own?"A named list, plus a documented way to add yours. Vagueness here means it cannot act, only speak.
"Show me a call where it did something you did not script."They can pull one up, because they read transcripts too. If every demo is the same demo, there is a script under it.
"What does it do when it is not confident?"A described behaviour, not a promise that it always is. Confidence handling is the single clearest signal of a real implementation.
"Where does my data go, and where is it processed?"A straight answer about which components run where, which models are used, and what is retained. Vagueness here is the one that later becomes a privacy problem.
"What is your containment, and what is repeat contact alongside it?"Both numbers, from real deployments. A vendor who has never been asked the second one is telling you something.

Then ring the demo number yourself and go off script. Interrupt it mid sentence. Change your mind halfway through. Give it a suburb and a vague description instead of a reference number. Ask for a person. Say something with two requests in one sentence. Every one of those is ordinary customer behaviour and every one of them separates an agent from a menu wearing a better voice. Our related piece on what sits underneath an AI provider covers the infrastructure side of the same evaluation.

How We Do It

Uniden Voice over Cloud runs the agent inside the phone platform rather than beside it, which matters for three specific reasons. The agent answers on the actual call path, so there is no forwarding hop, no second provider in the audio, and no delay while a call is handed between systems. It has the call context from the first second, including the number, the matched customer record and the history, because the platform already holds them. And it escalates into the same ring groups, queues and on-call rotations the rest of your call flow uses, so escalation is a route rather than an integration.

We build the first one with you on the shape described above: one narrow job, written guardrails, a defined escalation, a fortnight of shadow running where nothing reaches a customer, then after hours, then overflow. You get the transcripts from day one, the five numbers weekly, and a named person on our side who reads them with you. The platform is Australian owned, hosted in Australia and supported from Australia, and the APIs are open, so what you build with us is yours to connect to whatever else you run.

Start with one job, not the whole phone

Tell us what your after hours calls look like and we will scope a first agent around them: the job, the guardrails, the escalation and the success test, with two weeks of shadow running before a single customer hears it.

Get Started Or call 1300 881 662

Frequently Asked Questions

What is the difference between an AI agent and a chatbot?
A chatbot matches what it hears against a list of expected inputs and returns a prepared response or a route, so when the caller says something that is not on the list it fails, loops or falls through to a default, because there was never anything behind the list. An agent is given a goal rather than a script, decides which steps to take, calls tools to look things up and change things, and judges when it is finished or when it should stop and hand over. The practical test is whether the thing can do something it was not explicitly told to do, in service of a goal it was told to achieve. If a caller says "I rang yesterday about the leak at the Brunswick site and nobody has been out", a menu hears none of that, a fixed automation was not triggered by it, and an agent can find the job, check whether the technician attended, and either explain or escalate with the context intact. A third category sits between them: automations, which run a fixed sequence when a trigger fires, such as sending an SMS after a missed call. Those are genuinely useful and much simpler, and a great many businesses would get more value from three good automations than from an agent they have not scoped properly.
Why do so many AI agent projects fail?
Gartner expects more than forty per cent of agentic AI projects to be cancelled by the end of 2027 and names three causes: escalating costs, unclear business value, and inadequate risk controls. None of them are technical, and all three are decided at scoping time rather than during the build. In practice the pattern is consistent. The first job chosen was too big, usually the complicated one that eats a senior person's week, which is low volume, high variance and high consequence, meaning there is never enough repetition to tell whether it is working. No success sentence was written with a number and a date in it, so a year later nobody can say whether it worked and the project dies of ambiguity rather than failure. Guardrails were written as guidance instead of prohibitions, and enforced by instruction instead of by withholding the capability. Escalation was built as an afterthought, so the calls that mattered most were handled worst. And nobody read the transcripts, so the thing never improved. Businesses that get value run many small agents with narrow jobs rather than one with a broad one: the average Australian organisation using AI runs around eleven at once.
What job should my first AI agent do?
Pick one with high volume, low variance and low consequence, and accept that it will feel unambitious. High volume, because you need at least a few dozen instances a week to generate enough examples to judge it. Low variance, because agents are impressive in demos and inconsistent in production when the work has many shapes. Low consequence, because your process for catching and fixing mistakes does not exist yet and you want to build that habit somewhere forgiving. For most Australian businesses between five and two hundred staff, five jobs pay back inside a quarter: answering after hours calls that currently reach a mailbox, booking and rescheduling, job and order status enquiries, writing call notes and updates back into the CRM, and taking overflow at peak so calls are not abandoned. Four of those five are work nobody is currently doing, which matters more than it looks. An agent that picks up unhandled work does not require anybody to give anything up, so it gets a fair trial. An agent introduced as a replacement for a person's work gets audited by that person, and they will find the three calls it got wrong long before they find the ninety it got right.
What guardrails does an AI agent need?
Write them as flat prohibitions with no conditions attached, because guidance gets interpreted and prohibitions get enforced. The standard seven for a phone-connected agent: no pricing beyond the published rate card, since a quoted price on a recorded call is a representation and under Australian Consumer Law it is the business's regardless of who said it; no commitments about timeframes, because "someone will be there this afternoon" is the most common over-eager promise and the one that generates complaints; no account changes without verification, with a defined failure path that is not simply an easier second question; no discussion of any other customer or job; no claim to be human, and preferably disclosure up front, since businesses that disclose get fewer complaints than those whose customers work it out; no handling of a caller in distress, with defined trigger words routing straight to a person; and no unbounded retries, meaning two attempts to understand and then escalate. The most important structural point is to enforce guardrails in the tools rather than only in the instructions. An instruction not to issue refunds is a preference. Not giving the agent a refund tool is a control.
How do I measure whether an AI agent is working?
Five numbers, and the fifth is the one most dashboards leave out. Containment, the share of conversations finished without a person, which for a well-scoped first job settles somewhere between forty and seventy per cent; anything close to a hundred means the scope is trivial or escalation is broken. Resolution, the share of contained conversations where the customer's actual purpose was achieved, which requires reading a sample of about fifty calls a month rather than trusting the log. Escalation quality, the share of handovers where the person who took over had the context and the customer did not have to start again, which should be above ninety per cent and is a configuration problem when it is not. Cost per resolved contact, including the supervision time, because a comparison that excludes review hours is dishonest and gets caught at the first budget review. And repeat contact within seventy-two hours, the share of contained conversations where the same customer comes back about the same thing. That last one is decisive: if containment rises while repeat contact rises, the agent is deferring work rather than resolving it, and you are paying for two contacts instead of one. Report containment and repeat contact as one line, never two.
Is there a law in Australia about using AI agents in business?
There is no standalone Australian AI Act and, as things stand, no economy-wide mandatory guardrails. Ten mandatory guardrails for high risk AI were proposed in September 2024, and the National AI Plan of December 2025 confirmed Australia would instead rely on existing laws and sector regulators supported by voluntary guidance and the Australian AI Safety Institute. In July 2026 the Government set out a further direction including legislating Australian Standards for AI and creating an Office of AI within the Department of the Prime Minister and Cabinet, so the position may move. The expensive misunderstanding is to read that as nothing applying. Privacy law, the Australian Consumer Law, record keeping obligations, industry codes and the telecommunications rules all apply to what your agent says and does exactly as they apply to a staff member, and a misleading statement is not less misleading for having been generated. One dated change matters directly: from 10 December 2026 privacy policies must disclose the kinds of personal information used in substantially automated decisions and the kinds of decisions made, which reaches any agent that decides something about a person. Separately, mandatory requirements including AI impact assessments and Chief AI Officers began applying to Commonwealth agencies from 15 June 2026, with full compliance due 10 December 2026. Those bind agencies, not private businesses, but they make a reasonable template.
How long does it take to deploy an AI agent?
Ninety days for the first one if it is done properly, and about a fortnight for the second, because by then the connections, the guardrail habits and the review routine already exist. The sequence runs in four phases, each with an exit test rather than a date. Days one to fourteen are shadow running: the agent works on real calls and produces what it would have said and done, but nothing reaches a customer and no tool it calls changes anything, while a person reads a sample daily. You exit when you have read about a hundred transcripts and can predict its behaviour. Days fifteen to thirty-five are live but after hours only, on calls that would otherwise reach a mailbox, so the downside is bounded because the alternative was nothing. Days thirty-six to sixty are supervised daytime overflow, taking only calls that would otherwise queue past a threshold, with a named person spending half an hour a day on review. You exit when containment is stable across three weeks and repeat contact is flat or down. Days sixty-one to ninety put it on the front line for the defined job only, widening autonomy one decision at a time. Three things run throughout: the team is told before launch and given a one click way to flag a bad call, one named person owns the daily review inside their workload rather than on top of it, and every change is logged with a date.

What to Read Next

Your next reads

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