AI is expensive. You already knew that.
What you may not have fully priced in yet is that AI is now making your MacBooks expensive too.
This week, Apple confirmed what supply chain watchers had been warning about for months: memory prices have risen so sharply that even the world’s most powerful supply chain can no longer absorb them. MacBook Air prices jumped $200. MacBook Pro, $300. And Apple has signalled this isn’t the last move; prices may continue rising into 2027.
The headlines are framing this as a consumer story. A bad moment for people upgrading their personal laptops.
That framing is missing where the actual cost lands.
For companies with distributed global or large workforces, this isn’t a consumer inconvenience. It’s a hardware budget crisis arriving in the middle of a fiscal year that was planned using last year’s numbers.
The procurement problem nobody is talking about
Most corporate hardware budgets are set annually. Device refresh cycles are planned months in advance. Provisioning costs for new hires are estimated at the start of the year and locked in.
None of those estimates accounted for a 98% increase in DRAM pricing in a single quarter.
The IT manager who approved a provisioning budget in January is now looking at a $200–300 gap per device; before any regional pricing adjustments, before currency exposure, before import duties, before VAT.
For a company onboarding 50 new employees this year, that’s a six-figure shortfall in a budget line that was already considered “handled.”
The CFO conversation that follows is not a comfortable one.
The global multiplier
For companies with employees across multiple countries, the Apple price increase is not a flat $200. It’s a starting point that regional economics then amplify.
In markets where Apple hardware is imported, base price increases compound through distributor margins, import duties, and local tax regimes. In countries with currency exposure to the US dollar, exchange rate movements add another layer. In regions where Apple’s authorised distribution is thinner, availability constraints mean companies sometimes pay above-list for devices they need to ship in a specific window.
A $200 increase in San Francisco becomes a materially different number by the time it reaches a new hire in Manila, Lagos, or Bogotá.
For IT teams managing provisioning across six countries; a normal configuration for any mid-market company with remote-first or global hiring ambitions: the per-device cost variance is now significant enough to affect hiring timelines, onboarding schedules, and device standardisation strategies.
The hidden exposure: devices that haven’t come back
This pricing shift creates a new urgency around something most companies handle poorly.
Every device that doesn’t get recovered after an offboarding is now more expensive to replace. Every laptop sitting in a former employee’s home office or lost to a country where you have no recovery process, or flagged as “pending return” for the past six months: that’s $1,299 to $1,999 sitting outside your inventory.
This was always a cost. It is now a significantly larger cost.
For IT teams that have clear, automated offboarding workflows and real-time device inventory; this is a manageable problem. You know where every device is. You know which ones need to be recovered. You can act quickly.
For IT teams running device tracking on a spreadsheet or relying on manual processes that break down under pressure; the gap between “devices we think we have” and “devices we can account for” just got more expensive.
The BYOD temptation
When hardware gets expensive, companies start looking for shortcuts. The most common one is BYOD: Bring your own device; which shifts the hardware cost off the corporate balance sheet and onto the employee.
This is not a new conversation. But it tends to accelerate when device costs rise sharply.
The operational consequences are real: inconsistent security configurations, no MDM coverage, no ability to remote wipe, no clean offboarding path, and a shadow IT problem that expands every time someone connects a personal device to company systems.
Trading a hardware cost for a security exposure is not a neutral trade. But it’s a trade that looks more attractive to a finance team reviewing a hardware budget that no longer matches the year’s actual costs.
IT teams should expect this conversation. They should be prepared for it.
What this means for the next 18 months
The memory situation is not resolving quickly. New manufacturing capacity for traditional DRAM takes years to come online. As long as AI infrastructure investment continues at its current pace; and there are no credible signals it won’t: memory suppliers have limited incentive to prioritise consumer device allocation over high-margin data centre orders.
The companies that manage this period well will be the ones that have already built the operational infrastructure to make every device count:
Clean, real-time device inventory; not an annual audit, but a live record of every piece of hardware, its status, its location, and who is responsible for it.
Automated offboarding that triggers device recovery at the moment employment ends; not three weeks later when someone finally sends the return shipping label.
A clear BYOD policy with the security controls to back it up; so that when the finance team asks, IT can present a governed alternative rather than a reactive no.
Regional pricing awareness built into provisioning planning; so that the next hardware budget reflects what devices actually cost in each market, not what they cost in headquarters.
These are not new requirements. They’ve always mattered. They matter more now.
The AI boom is being covered as a software story, an infrastructure story, a jobs story.
It is also a hardware story. And the hardware story is landing on the desk of every IT manager who has to explain to a CFO why the laptop budget needs to be revised, again and what they’re going to do about the 40 devices that still haven’t come back from the last round of offboarding.
That’s not a consumer problem.
That’s an operational one.
