The real story behind Apple's AI announcements is friction
Why Apple's AI moves matter to higher ed
Apple's recent announcements (https://www.apple.com/newsroom/2024/06/introducing-apple-intelligence-for-iphone-ipad-and-mac/) about AI caused considerable stir among those of us interested in AI. The long term plan is for AI to be embedded in pretty much everything you do on your phone ... eventually. For example, AI will scan your inbox, prioritize messages, and draft replies while also adding events to your calendar and to-do list. (Again, I think this is the long term vision. These functions aren't available today.)
Initially, it looks like much of this integration will be through Siri, but AI will also be baked into apps such as Mail, Messaging, and Notes, and will also make for more effective notifications. There will also be a writing assistant that will do things like summarize long message threads, draft messages, and transcribe phone calls.
The Siri integration will be a big deal. In its current state, Siri is hit-or-miss at best. That's about to change. Siri will soon be MUCH more capable. For example, Siri will be able to do things like scanning a photo of a business card for an address and then feeding that address into the Maps app. Apple is also releasing an API (application programming interface) that will allow third-party developers to tap into the Apple AI ecosystem. This will offer up many other possibilities for AI app integrations by allowing non-Apple developers to integrate their apps into Apple's AI.
There are a bunch of other features. Check out the announcement for more details. Apple Intelligence (leave it to Apple to try to co-opt a well-established acronym like AI) should be available in the fall, but only for some devices. My BRAND NEW iPhone 15 isn't on the list. Sigh. iPhone 15 Pro and Pro Max are on the list, as are most iPad and Macs with M1 or later CPUs.
What’s not being said
Here's what most pundits are missing. The real news here isn't about the new features, it's about the reduction of AI friction. Huh? What does that mean? Friction is basically something that inhibits free movement. Friction can be friend or foe. Just ask anyone who's slipped on ice wether they would have liked some more friction or who has had an engine seize up due to a lack of friction-reducing lubrication.
When it comes to technology, friction is the overhead that users face when trying to actually use a technology. If you've ever had to search around for the right menu to do some task, you've experienced technology friction. It's safe to say that anyone who uses information technology experiences a lot of friction. In fact, one of the main goals of user interface design is to reduce friction.
Generative AI, in its current form, has a lot of friction. That's why prompt engineering is a thing. It's still pretty hard to get generative AI to do exactly what you want. Trying to use AI to integrate across apps is basically out of reach for anyone without specialized expertise. Apple Intelligence has the potential to change that. If it lives up to its promise, AI will just work on Apple devices. THIS IS HUGE (if it comes to be).
Only Apple
Apple is in a unique position to pull this off because it's ecosystem is so tightly integrated. I'm typing this on a Mac Mini Pro. I can copy something from this computer and paste it on my iPhone or iPad. My Apple Watch unlocks the Mac Mini and iPhone. I can reply to Messages on my phone, tablet, watch or computer. It's all pretty amazing. Apple devices just work well together. Apple also has a lot of control over hardware, especially since they started making their own chips. Google and Microsoft simply cannot match that level of integration. If you're not an Apple person, you probably have an Android phone and tablet and Windows PC (or maybe Linux). It's likely that some of your devices come from different manufacturers. This lack of integration makes truly embedding AI throughout the ecosystem extremely difficult.
So, in the near future, millions of Apple users will have significant AI capabilities at their fingertips (or more likely voice) with very little in the way of a learning curve. Friction won't be entirely eliminated, but it will be vastly reduced. This, I believe, has the potential to make the use of AI explode in a way we haven't experienced yet. It may also make the lives of Apple users better.
Of course, the picture isn't all sunshine, rainbows, and unicorns. Apple is talking big, but the reality may be different. (Although Apple's track record on delivering is better than most.) There are also huge privacy concerns, despite Apple's efforts to allay those fears. AI also screw up, just like real assistants, although errors should get better over time. As I often say, we'll see. But I definitely think that we're on the brink of a significant new infusion of AI into everyday life.
What does this mean for higher ed? It means that we'll see significantly increased use of AI. That's the bottom line. This will be especially true in the USA, where iPhones are more widely used than Android phones. So, like it or not, the odds are that we'll be dealing with more and more students using AI, which is yet another reason to develop sound policies around AI use and to modify our assignments to fit an AI world. As the saying goes, "You ain't seen nothing yet."