What's possible with AI in sales
Evidence current to mid-2026.
In sales, AI is ready today for the writing and the admin, like call notes and summaries, customer relationship management (CRM) updates, drafting outreach and prepping for meetings, always with the rep reviewing. The price, the close, the forecast you commit to and the claims you make to customers stay with people. The tables below show exactly which is which.
Each task below sits in one of three bands: a strong fit today, worth a careful pilot, or keep with people for now.
A clear breakdown of where AI does and does not fit across the core tasks of a sales team. A practical starting point, not the last word.
Where AI is mature: strong fit today
Proven and available today. AI does the bulk of the work and the rep reviews it.
| Function | The job today | With AI |
|---|---|---|
| Call and meeting notes and summaries | Take notes during the call, write them up, capture actions and next steps | AI records with consent and drafts the summary and actions; the rep checks attribution and owns what goes in the CRM |
| CRM updates and activity logging | Log calls, emails and meetings, keep contacts and stages current | AI captures activity from email and calendar and drafts the updates; the rep owns stage changes and keeps the data clean |
| Outreach and follow-up drafting | Write the first email, the follow-up and the recap, tailored to each person | AI drafts from a short brief and your notes; the rep edits for truth and tone and owns the send |
| Sales content and collateral | Write battlecards, one-pagers, email templates, call scripts and case-study drafts | AI drafts from a brief and your material; the rep checks the facts, the competitive claims and the brand voice |
| Account research and meeting-prep briefs | Dig through the website, news, LinkedIn and filings before a meeting | AI pulls the public information into a first-draft brief; the rep verifies the names and facts before the meeting |
Where AI is emerging: consider piloting with a human gate
Promising but not yet proven at this scale. AI assists and the rep stays in the loop, so trial it on a contained scope first.
| Function | The job today | With AI |
|---|---|---|
| Prospecting and lead-list building | Research accounts and contacts, spot trigger events, build and enrich the list | AI researches and enriches at speed; the rep verifies the contact data and consent before anyone is approached |
| Inbound lead qualification and routing | Answer the enquiry, qualify against the ideal customer profile (ICP), route to the right rep, follow up fast | AI replies instantly, handles the logistics and routes; the rep takes the real qualifying conversation |
| Proposals, quotes and statements of work (SOWs) | Pull the requirements, assemble the proposal, set price and scope, route for sign-off | AI assembles the draft from your template and CRM; the rep owns the price, the scope and the commitments |
| Request-for-proposal (RFP) and security-questionnaire responses | Read each question, find past answers and evidence, draft, route to the expert | AI drafts from your approved library; an expert verifies every claim about capability, security and compliance |
| Pipeline inspection and deal-risk flagging | Review the pipeline deal by deal, spot what is stalling, decide what to chase | AI surfaces stalled deals and engagement signals; the sales manager judges the deal and owns the call |
| Conversation intelligence and coaching | Listen back to calls, score them, coach the rep | AI transcribes and flags talk-time and structure and points to the calls worth reviewing; the sales manager owns the coaching |
| Forecast roll-up and deal scoring | Roll up the reps’ commit calls, weigh the pipeline, build the forecast view | AI assembles the roll-up and flags risk in the numbers; the sales manager owns the committed forecast |
Where AI is not ready or suitable today: keep with people
These stay with people, either because the tooling is not reliable enough or because controls and compliance rule it out.
- Final pricing and discount authority. Setting the price and approving the discount is a commercial commitment. AI assembles the quote; the sales manager or owner sets and approves the number.
- Negotiation and the closing relationship. The trust that closes a deal is human, buyers still turn to a sales rep to validate what they have read, and selling people know is automated erodes trust. AI prepares; the rep closes.
- The committed forecast. The number you stake your credibility on to the board. AI can draft the roll-up, but predictive accuracy collapses on small or messy pipelines, so the commitment stays with the sales manager.
- Claims and commitments to customers. What you promise on price, capability, delivery or compliance is binding, and the business is liable for what its AI says (the Australian Consumer Law and the chatbot-liability cases are clear). The rep owns every claim.
- Mass autonomous outreach. Letting AI source, write and send cold outreach at scale, unreviewed, is where the “AI sales rep” promise broke in 2025: poor results, deliverability damage, and Spam Act and Do Not Call exposure. The rep owns who is contacted and what is sent.
Worth weighing: for any of these, the upfront setup (loading your templates, ICP, product and pricing rules, and agreeing the review steps) is a one-off, separate from the ongoing effort, which on the strong-fit items is mostly review. The time that comes back goes to the selling: the conversations that build trust, the research that sharpens the pitch, and the deals that need a real read of the room. AI takes the notes and the drafts; the rep stays where they move the deal.
The evidence
The research, regulators and tools behind this guide:
- MIT: 95% of enterprise AI pilots returned nothing (2025)
- Gartner: 69% of B2B buyers turn to a sales rep to validate what AI told them (2026)
- TechCrunch: the AI sales-rep startup that listed customers it didn't have (2025)
- ACMA: Commonwealth Bank pays $7.5m for spam breaches (2024)
- ACMA: Lululemon penalised $702,900 for sending promotional emails without an unsubscribe option (2026)
- ACCC: Microsoft in court over how an AI feature and price rise were presented (2025)
- OAIC: privacy and the use of commercially available AI products
This guide sits on top of the things that stay the same whatever your function. See the ground rules
Questions
The questions leaders ask.
Will AI replace my sales team?
No, it changes what they spend time on. The note-taking, the CRM admin, the first-draft writing and the meeting prep shrink, and their judgement, their conversations and the relationships they build matter more. The work that carries the deal, the price, the negotiation, the close and the forecast you commit to, stays firmly with people. The goal is a sales function that does more with the same people, not fewer.
Is it safe to use AI with customer data and in front of customers?
It is safe when you control where the data goes and the rep stays in the loop. Keep processing in Australia where you can, turn data retention and training off, and put in only the customer detail the task needs. In front of customers the rule is firmer: the business is liable for what its AI says on its behalf, so the rep owns every claim, and outreach has to meet the Spam Act and the Do Not Call rules. From 10 December 2026, if you use automated systems to score or route leads, you are likely to need to disclose that in your privacy policy under the Privacy Act amendments; run an audit now. Used that way AI drafts and researches while the commitments stay human.
Where should we start?
Pick something repetitive and easy to check. Call notes and summaries are a common first step: AI records with consent and drafts the summary and the actions, and the rep confirms what goes in the CRM. Drafting outreach and follow-ups is the other strong one, because the blank page is the real chore and the rep still edits for truth and owns the send. Prove one before adding the next.
Do we still need salespeople?
Yes. The work shifts from admin and drafting to judgement and relationship, and the moments that close a deal stay human. Independent research is clear that buyers still turn to a sales rep to validate what AI told them, and that selling people know is automated erodes trust. AI helps the rep prepare and follow up faster; it does not build the relationship or own the commitment.
What can AI do in sales now that it could not six months ago?
Quite a bit, which is why this is dated. Through early 2026 the meeting recorders and CRM assistants matured, AI account-research agents made pre-meeting prep far faster, and conversation-intelligence tools became standard. The honest other half of the story is that the fully autonomous "AI sales rep" promise did not hold up, with the most-hyped tools posting poor results and real trouble. A year on, the assisted tools are usable, with the rep still checking the output and owning the deal.
How current is this, and what if the tools have moved on?
The evidence is current to mid-2026 and we refresh it as the tools and the rules change, which in sales they do often. Where a finding comes from an independent source we lead with it; where it comes from a vendor we treat it as marketing. If something has shifted since you read this, the fastest way to get the current picture for your own stack is a quick chat.
Book a discovery callA first conversation
See how this applies in your world.
Book a confidential chat. Tell us where you see the opportunity, in sales or anywhere else, and we will agree a sensible next step. No deck, no pitch.