Written before the event — This article was published on 24 February as a forecast for Apple's 4 March event. That date has long passed, so what follows is a record of what was expected rather than what was announced; check Apple's own newsroom for the products that actually shipped. How to read an Apple rumour — which parts of a leak are reliable, which are routinely wrong, and why — follows the original article.
Apple has said nothing officially, which has never stopped anyone. The company’s “Apple Experience” event on 4 March has become the focus of a familiar guessing game, and the most reliable player in it — Bloomberg’s Mark Gurman — expects a run of announcements starting on 2 March, followed by hands-on time for journalists on the 4th. Reports across the international press converge on at least five products. Here they are, with a view on what each means for buyers here.
iPhone 17e
The successor to last year’s iPhone 16e is expected to move to the A19 chip, bringing faster graphics for games and headroom for on-device AI features. Rumoured additions include MagSafe charging, Apple’s own newest cellular and wireless chips in place of third-party parts, and an 18-megapixel front camera with Center Stage, which automatically widens the frame for group selfies. The “e” line is Apple’s entry point, and in a market like Bangladesh — where iPhones arrive through grey channels at a premium — it is the model most likely to be the first Apple phone for many people.
iPads
The base iPad is tipped to get the A18, which matters for one reason: it is the floor for Apple Intelligence. The iPad Air is expected to move to the M4, and the iPad mini, while keeping its design, may finally get an OLED display.
MacBook Air and MacBook Pro
Updated MacBook Air and 14- and 16-inch MacBook Pro models are expected on the M5 chip, reportedly alongside Apple’s N1 networking chip for Wi-Fi and Bluetooth. For most people this is an incremental year; for people editing video in Dhaka on a five-year-old Intel Mac it is the upgrade cycle finally worth waiting for.
The one to watch: a cheaper, smaller MacBook
The most intriguing rumour is a MacBook with a screen smaller than 13 inches, running an iPhone-class processor rather than an M-series chip, priced to compete with Windows laptops and Chromebooks. Apple has never played in that price band with a Mac. If the reports are right, the device could do for laptops what the iPhone SE did for phones: bring students and first-time buyers into the ecosystem at a price that undercuts the resale-market MacBook Airs that currently fill that role in Bangladesh.
A note on rumours
Gurman’s track record is strong but not perfect, and Apple routinely shifts products between events. Treat the list as a well-sourced forecast: the M5 Macs and iPhone 17e are close to certain, the cheap MacBook is the one with the most room to disappoint — or to surprise.
How to read an Apple rumour
Forecasts like the one above are a permanent feature of the technology calendar, and they are more readable once you know where the information comes from. Leaks are not one thing. They arrive through three different routes, and the route determines which parts to trust.
Route one: the supply chain
Most hardware detail originates with the companies that make the parts. A phone requires display panels, camera modules, batteries and casings ordered months ahead from dozens of suppliers across several countries, each with its own staff, contractors and logistics.
That produces a reliable pattern. Specifications leak accurately and early, because a component order is a physical fact — somebody is tooling a production line for a particular screen size, and that is difficult to hide from the people operating it. Analysts who track these orders are often right about what a device will contain long before anyone has seen one.
The same route is unreliable about everything else. A supplier knows what it is building; it does not know what the product will be called, what it will cost, which features will be enabled in software, or whether the whole thing will be cancelled. Component orders also get placed for products that never ship, because planning requires committing to parts before committing to a launch.
Route two: the company itself
Some reporting comes from inside, and it is categorically different. Journalists with sources among employees learn about plans, internal timelines and decisions still being argued over — which is why this route is better on timing and strategy and no better than the supply chain on component detail.
It also carries a bias worth knowing. Companies brief selectively. A story that lands a week before an announcement, flattering and precisely accurate, is frequently not a leak at all; it is a soft launch, used to test reaction or to set expectations before a price. That does not make it false. It makes the absence of such a story informative in a way the presence of one is not.
Route three: everyone else
The third category is people with no access who are rewarded by attention for being first. Their claims are unfalsifiable until the event, cost nothing to make, and are occasionally right by chance — which is enough to build a reputation on if nobody keeps score.
The check is simple and almost nobody applies it: look for the prediction that was wrong. Any source with a long record has made mistakes, and the ones worth reading say so afterwards. A track record with no misses in it is not a good record; it is an incomplete one.
What is routinely wrong, in order
Across years of these cycles the error pattern is consistent, and it is useful for weighting any individual forecast.
Existence is usually right. If several independent reports say a device with a particular screen size is coming, it generally is. The parts are already being made.
Timing is often wrong. Products slip between events routinely, for reasons ranging from a software feature not being ready to a supplier missing a yield target. A device predicted for spring arriving in autumn is the commonest failure in this genre, and it does not mean the original report was fabricated.
Names are frequently wrong, because they are decided late and by marketing rather than engineering, and the people who know the hardware are not the people who know the name.
Prices are wrong more often than anything else. They are set last, they depend on currency and tariffs at the moment of launch, and they are the detail a company guards hardest because it determines the first day's coverage. Treat any pre-announcement price as a guess wearing a number.
The category that deserves the most scepticism
One pattern is worth isolating because it recurs every cycle: the cheaper device in a price band the company has never entered.
These rumours are perennial, they are the most exciting, and they have the worst hit rate — partly because such a product would cannibalise a profitable line, and partly because "they might make a cheap one" is the easiest plausible thing to say about any premium manufacturer. Sometimes it happens. It happens far less often than it is predicted, and a forecast is at its weakest precisely where it is most appealing.
The practical position
None of this means ignoring pre-event coverage, which is often genuinely well sourced and is the only information available. It means reading it with the weights attached: near-certain on what parts exist, loose on when, unreliable on what it will be called, and close to worthless on what it will cost.
And the buying rule that follows is the same one that applies to every announced-but-unshipped feature: a product is worth what it does when you can buy it. Waiting for a rumoured device is a decision to be without a working machine for an unknown period, in exchange for a specification that may change and a price nobody has confirmed.
