This has been the most frequently asked question from our clients and candidates for the past 18 months. It’s easy to focus on given the economic data and the steady flow of headlines about mass layoffs, the latest advances from labs like Anthropic, and warnings of economic downturn – they seem to be everywhere we look. However, the current reality is quite different.
We’ve seen nearly every industry and sector awash with AI automation platforms and agentic tools, from real estate fund reporting to legal contract drafting to automated payroll services. Identifying which platform, or which set of AI agents, to introduce to a firm is a genuinely complex decision for most business leaders. The considerations tend to be:
Is this platform truly appropriate and applicable to our sector?
Where’s our competitive advantage in adopting it? Is it cost-effective?
How do we make the most of it while still protecting collaboration across teams?
Are there specific features better suited to one client base than another?
And, ultimately – how will this affect the existing team and future hiring?
MIT’s NANDA initiative, in its “State of AI in Business 2025” report released last year, drew on 150 leadership interviews, a survey of 350 employees, and an analysis of 300 public AI deployments. It found that only about 5% of AI pilots achieved rapid revenue acceleration, with the large majority stalling and delivering no measurable P&L impact. Notably, though, tools purchased from specialised vendors succeeded roughly 67% of the time – around double the success rate of AI systems built in-house.
On workforce impact specifically, the report’s conclusion was clear: generative AI is already affecting the workforce, but mainly through selective displacement of functions that were previously outsourced, and through more constrained hiring, not through broad-based layoffs. The report frames the real barrier to scaling AI not as regulation or a lack of capable technology, but as a “learning gap” where generic off-the-shelf tools don’t adapt to a firm’s specific workflows, and most enterprises struggle to integrate them effectively.
Take a typical Investment Analyst role as an example:
| Task Area | Before AI | Post AI | Freed-up capacity → increased opportunities |
| Financial modelling | Analyst builds models from scratch, manually links formulas, checks for errors | AI drafts/updates model shells, flags inconsistencies, runs sensitivity scenarios | Time to stress-test strategic assumptions (macro scenarios, management credibility) rather than mechanics – moves the analyst from model-builder to sense-checker |
| Company/industry research | Pulling and assessing filings, transcripts and news into a memo | AI summarises filings, transcripts and comparable-company data | Bandwidth for client-facing work: more time with perspective clients and to conduct account management to existing relationships. More site visits and channel checks, and overall more time building qualitative judgement |
| Data gathering & screening | Manual screens across databases (Bloomberg, CapIQ) for comparables or targets | AI runs and refines screens, flags anomalies | More time spent gaining exposure to deal origination – sourcing ideas rather than just processing pipeline handed down |
| Report/memo drafting | First-draft writing | AI produces first drafts | More time in investment committee discussions – junior analysts historically wrote memos but rarely had a voice in the room; freed-up time can buy a seat closer to that table |
For the firms getting AI deployment right, this shift is freeing analysts to spend more time on relationship-building, client-facing work, and spotting new investment opportunities – moving their value add and now with greater contribution to origination.
Also, take a typical Financial Accountant position –
| Task Area | Before AI | Post AI | Freed-up capacity → increased opportunities |
| Financial reporting | Significant time spent compiling reports and formatting | AI assists with drafting management reports, commentary and presentation material | More time interpreting results and communicating insights to leadership |
| Data extraction | Exporting data from multiple systems into Excel | AI retrieves and consolidates data automatically from connected systems | Faster reporting cycles and more time for analysis |
| Budget & forecast support | Manual collection and manipulation of historical data | AI prepares baseline forecasts using historical and current data | More involvement in scenario modelling and strategic planning |
| Stakeholder queries | Responding individually to routine finance questions | AI answers common finance queries or retrieves information quickly | More time supporting senior stakeholders on complex commercial decisions |
Where AI has been implemented effectively, Financial Accountants are spending less time producing information and more time interpreting it, helping leaders make better-informed decisions.
The World Economic Forum’s Future of Jobs Report of 2025 (survey of 1,000+ employers across 55 economies) found that 77% of employers plan to prioritise upskilling/reskilling their existing workforce as their leading response to AI and other workforce disruption by 2030 – ahead of hiring new AI-skilled staff.
Whilst the cost of reducing headcount and then only to find in 18 months’ time, that the same skillset is still needed, additional AI training still has to be rolled out, and that a more considered read of the next few years, would have pointed the other way. The cost of this is immeasurable, as rehiring and re-onboarding is expensive, slow, and erodes institutional knowledge in a way that’s hard to price on a spreadsheet.
EMA (NZ) found that 83% of New Zealand businesses are using AI, yet only 13% have a formal AI policy. Perhaps most strikingly, just 4% of AI initiatives are led by HR, highlighting that many organisations are focusing on the technology before redesigning the work and workforce around it. Competitive advantage won’t come from adopting AI alone. It will come from making deliberate decisions about where AI accelerates work and where human judgement continues to add the greatest value.
It’s at this point that we have to remind ourselves why we hire people in the first place. Judgement calls under uncertainty, relationship-building, and the ability to read a room or a client’s real intent, are exactly what current AI tools can’t replicate and exactly what clients are paying a premium for in this market. The firms that treat their people as assets that can be adapted, rather than the first line item to cut – are the ones best placed to still have the right team in place when the dust settles. Ironically, AI should actually be making us more, human?