AI Just Wiped Out 10,000 Tech Jobs. Is Your Finance Function Next — or Actually Safe?
AI Just Wiped Out 10,000 Tech Jobs. Is Your Finance Function Next — or Actually Safe?
Labels: AI Finance | Finance Operations
The headlines this week were hard to miss. Meta announced it was cutting 8,000 jobs. Snap said AI now generates more than 65 per cent of its new code and proceeded to let 1,000 people go — citing $500 million in annualised savings. Microsoft quietly launched a voluntary buyout program targeting roughly 7 per cent of its US workforce. And Salesforce trimmed around 1,000 roles, with CEO Marc Benioff pointing to AI productivity gains as the reason smaller teams could now do more.
If you work in finance, you've probably seen these stories and wondered — even briefly — whether your function is next on the list. It's a reasonable question. And it deserves a straight answer rather than corporate cheerleading in either direction.
The short version: parts of the finance function are genuinely at risk. Other parts are becoming more valuable. Understanding the difference is the most important career and operational conversation finance leaders should be having right now.
|
~10,000
Jobs cut across Meta, Snap, Microsoft and Salesforce in the past month — with AI cited as a direct enabler in each case.
|
65%
Of Snap's new code is now generated by AI — a figure that jumped dramatically within a single product cycle.
|
|
$500M
Annualised cost savings Snap is targeting from its AI-driven restructure — a figure that's reshaping investor expectations across the sector.
|
~7%
Of Microsoft's US workforce targeted in a voluntary buyout program launched in late April 2026 — its first program of this kind.
|
What's Actually Happening in Big Tech
It's worth being clear about what these job cuts represent. In most cases, companies aren't replacing individual workers with an AI model doing the exact same job. What's happening is more structural: AI tools are increasing the output per person in certain functions to the point where fewer people are needed to achieve the same result. In engineering and coding-adjacent roles, that shift has been rapid and measurable. When 65 per cent of new code is AI-generated, the human input required shifts from writing code to reviewing, directing, and making judgement calls about what to build.
The finance implication isn't difficult to extrapolate. If AI can generate, reconcile, and format the repetitive outputs that used to require analyst hours, the question of how many analysts you need to employ starts to look different. That's not a future-state scenario. It's already happening in finance functions that have invested in automation.
What AI Is Replacing in Finance — Honestly
Finance professionals are sometimes reluctant to admit which parts of their work are genuinely automatable. But intellectual honesty here is more useful than defensiveness. The tasks AI is already replacing — or significantly accelerating — in finance functions include:
- Data extraction and entry — pulling figures from bank feeds, invoices, and reports and pushing them into structured formats.
- Standard reconciliations — matching transactions across ledgers, identifying and flagging discrepancies for human review.
- Report compilation — assembling management packs from data that already exists in structured form, formatted to template.
- First-pass variance commentary — generating an initial explanation of movement based on known drivers, for a finance professional to review and refine.
- Payroll exception flagging — identifying anomalies in payroll runs before they become errors.
None of these are new observations. Finance functions have been automating transactional work for decades — through ERP systems, bank feeds, and rule-based automation. Generative AI is accelerating the timeline and expanding the scope, but it is not inventing the trend. What it is doing is making automation accessible without requiring specialist technical skills to implement.
What AI Cannot Replace — and Why Finance Is Safer Than It Looks
Here is where the "finance teams are next" narrative tends to fall apart. The tasks that define the value of a strong finance function are not primarily about generating outputs. They're about interpreting them, contextualising them, and taking accountability for the judgements they support.
| What AI can do | What still needs a finance professional |
|---|---|
| Generate variance analysis from structured data | Know that the variance is actually driven by a management decision that isn't visible in the numbers |
| Compile a management reporting pack to template | Decide what the board actually needs to hear this month — and what to leave out |
| Flag potential compliance risks based on known rules | Interpret regulatory grey areas and make the call under uncertainty |
| Model financial scenarios based on provided inputs | Determine which scenarios are worth modelling given the organisation's strategic context |
| Draft financial commentary from data patterns | Take accountability for the narrative presented to a board or audit committee |
The pattern is consistent. AI accelerates the production of outputs. It does not replace the judgement, context, accountability, and relationship layers that give those outputs meaning in an organisational setting.
Finance professionals who operate predominantly at the output-production level — those whose primary value is in compiling, formatting, and distributing — face genuine displacement risk. Finance professionals whose value is in the interpretation, advice, and accountability layers do not. In many cases, they become more valuable as AI handles the production work, freeing them to focus entirely on the judgement work.
The real risk isn't being replaced by AI. It's being replaced by a finance function that uses AI effectively, and is therefore able to do more with a smaller team. For finance leaders, the question isn't "will AI take my job?" It's "am I upgrading my function fast enough that we stay competitive on output and cost?"
What This Means for Finance Leaders Right Now
The big-tech restructures are a useful reference point, not a direct prediction. Australian SMEs, NFPs, and NDIS providers operate in very different cost environments to Snap or Meta. They're not going to fire their finance teams because AI wrote some code. But the underlying dynamic — AI enabling smaller teams to deliver the same output — does apply.
Finance leaders who are building AI-augmented functions now are creating a structural advantage: the same quality of financial management at a lower ongoing cost, with faster turnaround on reporting and analysis. Finance leaders who aren't are building a cost structure that will increasingly look out of step with peer organisations that have moved.
The first question isn't "which tools should we implement?" It's "which parts of our current finance function are genuinely adding judgement value, and which parts are producing outputs that could be systematised?" That audit — honest and unsentimental — is where the conversation needs to start.
Not Sure Where Your Finance Function Stands?
PFL works with CFOs and finance leaders to map where their teams are genuinely adding value — and where automation can take the weight. If you're thinking through what AI adoption looks like for your function, we're happy to start that conversation.
Talk to PFL →
Comments
Post a Comment