AI in the dealership, minus the hype: what it changes, what it can't, and where you stay in charge

AI is now in more Australian businesses than ever, but the sales pitch and the showroom reality are two different things. If you run a dealership, the useful question is not whether AI is coming. It is what it genuinely changes on your floor, what it still gets wrong, and where a person has to stay in charge.

Adoption is real, not hypothetical. According to the Australian Bureau of Statistics, around 12 per cent of Australian businesses reported using AI in their workplace in 2024 to 2025, and almost one in two (46 per cent) said they were finding savings and efficiencies through innovation and new technologies, including AI. So the trend is clear. What matters is putting it to work in ways that hold up.

What AI actually changes on the floor

Strip away the buzzwords and the practical wins are narrow but valuable. AI is good at the repetitive, time-sensitive work that quietly leaks deals: replying to an enquiry within seconds at 9pm, keeping a follow-up cadence running when the team is flat out, drafting a first-pass message, sorting which leads look hot, and logging the notes nobody enjoys writing.

None of that replaces a salesperson. It clears the runway for one. When the admin and the first response are handled, your people spend more time on the conversations that actually close: the test drive, the trade valuation, the finance chat, the handover. That is the honest version of the promise. AI does not sell the car. It makes sure the opportunity is still alive when your best person picks up the phone.

What it still gets wrong, and where a human stays in charge

Be clear-eyed about the limits. AI can state something confidently and be wrong. It cannot read a room, handle a delicate negotiation, or carry a customer through an emotionally charged moment. The moment a customer asks for a person, or the conversation gets complex or sensitive, the right move is a clean handover to a human, not a bot that keeps talking.

Customers feel this too. Research by KPMG and the University of Melbourne found that while half of Australians (50 per cent) use AI regularly, only 36 per cent are willing to trust it, and 78 per cent are concerned about negative outcomes such as inaccurate results and loss of human interaction. The same study found 65 per cent expect benefits from AI, and just over half (55 per cent) say they have realised them. Read together, that is your operating instruction: use AI for speed and coverage, keep a human accountable for judgement, accuracy, and the relationship. No tool can promise a sale, and no tool should be the last word on a customer's problem.

Consent, disclosure, and why being upfront pays off

Here is the part too many rollouts skip. The rules do not change because a machine sent the message. Under the Spam Act 2003 and Australian privacy law, consent, a working unsubscribe, and clear sender identification apply to an AI-sent SMS or email exactly as they do to one typed by a salesperson. Automating your outreach does not automate away your obligations. If anything, it raises the stakes, because a system can send at volume.

Disclosure, telling customers plainly that they are dealing with an automated assistant, is both good practice and good for trust, but it rewards being done well. It is also what Australia's Voluntary AI Safety Standard, published by the Department of Industry, Science and Resources, asks organisations to do: tell people when they are interacting with an AI system. Studies collected in the journal AI and SOCIETY show disclosing a bot's identity at the start can dampen engagement in some settings, including one field experiment where disclosure reduced purchase rates sharply, yet the same research found that pairing disclosure with context, such as what the assistant can and cannot do, softens that effect. Some customers also see undisclosed human-like bots as deceptive and say they value transparency. That lines up with the broader trust picture: the Melbourne Business School study found 83 per cent of Australians would be more willing to trust AI systems when assurances are in place, and research in Humanities and Social Sciences Communications found that transparency about how an AI works can reduce negative attitudes and lift trust in the company behind it.

So the compliant path and the smart path are the same one. Say it is an assistant, say what it does, make the human handover easy, and keep your consent and unsubscribe records clean.

Used this way, AI is not a revolution in your dealership. It is a quiet operational upgrade: faster first replies, follow-up that does not fall over, and more of your team's time spent where it earns its keep, all inside the rules you already answer to.

Sources: Australian Bureau of Statistics, Business adoption of Artificial Intelligence accelerates in 2024: 25, KPMG Australia and University of Melbourne, Global study reveals Australia lags in trust of AI despite growing use, Melbourne Business School (University of Melbourne), Trust, attitudes and use of AI: Australian Story, AI & SOCIETY (Springer), Understanding users' responses to disclosed vs. undisclosed customer service chatbots: a mixed methods study, Humanities and Social Sciences Communications (Nature), AI algorithm transparency: mitigating general negative attitudes and enhancing trust toward AI, Department of Industry, Science and Resources: Voluntary AI Safety Standard. Figures are industry findings, not Dealerloop results.