Apple Vision Pro - Seeing The World Through A New Lens

The Apple Vision Pro appeared on tens of thousands of doorsteps at the start of February. This is, in reality, a vanishingly small number: the closest comparable product on the market, Meta’s Quest, has sold tens of millions. But when Apple starts a new product line, the world takes note. Technically, this looks a big step forward compared to other VR/AR headsets, and demand was incredibly high in spite of the $3,500 price tag - which will of course come down over the coming years.

The big question for Apple, or for any of us thinking about what the future might look like, is: how many people will actually want one even when that price does come down? Is this the new Mac (~100m users), the new iPad (~250m), even the new iPhone (~1.5bn)? Or is it just a cool new gadget, thrown as a curveball into an AI-obsessed tech landscape, but with no clear long-term use case? Does Apple itself even have a view on this, or is it taking a punt - creating something new and seeing what uses developers and consumers will find for it?

Part of the hesitancy is because, at face value, the Vision Pro looks like a product of the pandemic. The demo videos present it as an awesome combined productivity and escapism tool for use at home (you are expected to be seated) and most likely in isolation (yes people can see your eyes through the glasses, but come on…). In a lockdown, unable to go out much and perpetually WFH, it would be the ideal device. But now, when we’re back out seeing friends and increasingly cajoled into physical offices? It’s hard not to feel as though the solutions it provides are to yesterday’s problems.

Spending some time thinking about it over recent weeks, however, the Vision Pro has started to make a little more sense. Oddly, this has been as a result of considering its capabilities through the lens of GenAI: framing the two not as separate technologies, but as complementary developments. Looking out around 5 years to an AI-enabled future of work, in particular, the Vision Pro could slot right in. On this reading, the Vision Pro isn’t an AR/VR play from Apple - or at least, it’s not only that. It’s a first bid to equip us with the tools we will want to be using in that future world.

A tool for a new kind of work

One of the more remarked-upon capabilities that the Vision Pro will bring to the workplace is the opportunity to do things in 3D. Frankly, no one quite knows what this means yet. But it could certainly transform data visualisation and analysis: making it possible to see the relationships between different variables in higher dimensionality, and better comprehend behaviours across networks and more complex systems. The way in which we try to understand and effectively communicate the complexity of the world we’re operating in, informing decisions across business, government and beyond, could change completely.

This is yet another indicator that the future of (white collar) work looks set to be dominated by ‘higher value’ activities. We’re already seeing evidence of the potential for GenAI to reduce the amount of time spent doing menial tasks, freeing up capacity for this higher value work. Some early studies have found significant increases in job satisfaction when using GenAI, purportedly on this basis. Other new gadgets aim to achieve something similar in our personal lives: the much-hyped Rabbit R1 promises to carry out tasks such as booking flights, making reservations or even editing photos with very little input, like a virtual extension of your will. The Vision Pro, by not trying to take work off your hands (by creating new types of work, even!) looks like it’s pointing in a different direction. But it’s actually part of the same picture: it’s just there to help you do more of that higher value work, and hopefully do it better.

Examples of how the Vision Pro might help include:

• Deeper work environments. High value, or ‘deeper,’ work typically requires greater focus over longer periods. In much the same way you might choose to listen to certain music to support this now, with the Vision Pro you’ll be able to change your surrounding environment, your whole perception of where you are, depending on what helps you work best, drowning out distractions around you in the process.
• Spaces for rest. High value work is also usually more fatiguing. Again, the Vision Pro seems set up for this, with the immersive environments for you to take a break in, seemingly ‘optimised’ for rest and mindfulness. And all just a few eye movements away.
• Attention monitoring. The eye tracking the Vision Pro depends on, along with the tracking of facial expressions for use by your avatar in virtual conversation, would also support attention and mood monitoring in future. This isn’t something which Apple has led on in its promo materials, perhaps unsurprisingly given the ethical issues it raises. Presumably it isn’t built into the current models. But there is a probable future in which at least some people want to be told when they should take a break, or have their environment change automatically to boost their flagging mood.
• Unlimited immersive training. Outside of a white collar context, the scope for training people in all manner of roles, at a much lower cost (you don’t have to physically recreate the environment and the people in it) and with personalisation for the user, seems incredible. Obvious examples include teaching, surgery, construction, or cooking.

Working with avatars

The Vision Pro also looks well set-up for a future in which we spend more time collaborating with intelligent agents, conversing with them as we go. The chat functionality where Vision Pro users can appear to each other as avatars, with the original person’s voice, facial expressions and hand gestures, could serve as a platform for conversing with AI-powered avatars, too. Depending on how the AI landscape develops (and/or Apple’s preferences and your own personal taste), you may just have one of these - it wouldn’t have to be a paperclip this time - or a whole cast of characters with different specialities, popping up to help you on a variety of tasks.

The Vision Pro would also be the perfect training environment for autonomous agent-avatars modelled on you, to be sent out into the world on your behalf. An autonomous avatar could feasibly learn your mannerisms from headset data - voice, turns of phrase, facial expressions, hand gestures, possibly more - and incorporate these into its presentation to others. It could represent you in (presumably low-stakes, at least for now) meetings, and report back. Similarly, you might find yourself having conversations with other people’s virtual avatar selves, instead of the ‘real’ avatars you might use the Vision Pro to speak to now…

This is very much in the realms of possible, rather than probable, future functionality. But it’s interesting to think about the ways in which the Vision Pro makes some paths more obviously plausible for the development of AI and our interactions with it - and on an accelerated timescale. A few months ago, it seemed obvious that we would have AI-enabled extensions of people carrying out tasks on our behalf in limited, impersonal contexts, and soon (writing for us, interacting with apps to make bookings and orders). But throw in the Vision Pro’s sensor-set and avatar chat, and this more personal version also becomes possible, expanding the range of contexts into which our AI-powered selves might extend before long.

-

It’s unclear how much of this possible AI-complementarity has been intended by Apple, and how much is us simply finding new uses for a fundamentally open-ended technology. There is little/no mention of AI in any of the materials released around the Vision Pro. But it also seems unlikely that these angles haven’t been considered, or even primed, ready for certain future directions of AI travel. No doubt there are many other possible angles we haven’t touched on here, too.

Either way, it’s interesting how the pairing of two technologies can help start to shed light on the possible futures of each. This is something we might explore again in future.

‍