Welcome back to Forests Over Trees, your weekly tech strategy newsletter. It’s time to zoom-out, connect dots, and (try to) predict the future.
Why are apps giving Siri new powers?
As kids, my brothers and I were not Michael Phelps.
But our mom wanted to make sure we knew how to swim. So even though we dreaded it every day, summer swim team was a must.
To keep our spirits up, we enjoyed a nice mixture of carrots and sticks. But carrots – like one video game for a summer of good swim team attendance – didn’t always cut it.
Sometimes, Mom had to bring out the sticks (extra chores, missed TV time, or a stern talking to).
But carrots and sticks aren’t just for bratty kids…
Apple is using them to supercharge Siri. I’ll explain… but first, let’s set some context.
Apple Intelligence
A few weeks ago, Apple announced “Apple Intelligence” at their annual keynote. Basically, they’re weaving AI into their software in several ways, while also re-branding the AI acronym as an Apple term (classic Apple!).
Even though it will only fully roll-out in 2025, people still went bananas.
Here’s how Apple Intelligence will change Siri:
Full in-app menu
Before, Siri could open a 3rd party app for you, but it couldn’t “do” anything inside that app. With this upgrade, you’ll be able to say “Siri, open Spotify, and add all of T-Swift’s albums to my breakup playlist”.
Multi-modal Siri
Siri is also embracing the multi-modal trend, so you’ll be able to interact using your voice plus the text you’re seeing on-screen. You’ll be able to pull up a restaurant’s reviews on Google Maps and ask “Siri, do any of these reviews mention cheesecake?!”
Actions without opening
Today, the stuff you can do using Siri without opening an app is pretty weak. You can start a timer, or ask about the weather, but that’s pretty much it. Now, without leaving your TikTok binge, you’ll be able to say “Siri, check me into my Delta flight.”
Spotlight Siri
Before, searching for something meant typing a search in Spotlight, or poking around in individual apps. With this upgrade, you’ll be able to say “Siri, where did Tim say he wanted to meet tonight?” and Siri can comb through WhatsApp, Messages, DMs, and email to find the right rendezvous point.
Natural language understanding
Last, but certainly not least, this upgrade will make Siri better at understanding you – even if you don’t phrase things perfectly. Hopefully, this will cut down on the number of times Siri responds to your perfectly phrased question with “Sorry, I didn’t catch that.”
Here’s a clip that shows-off some of those new powers:
Reacting to the news
As a user, this is cool as hell. It makes it easier to get things done, I’ll probably spend less time looking down at my phone, and I get to feel like I’m living in the future.
For Apple, this is also cool as hell. Unlocking magical experiences for users is always a good thing. Plus, they’re uniquely positioned to do it!
OpenAI and the model makers have pushed the industry forward, but they can’t do something quite like this (even if they start to “remember” your past GPT queries), as I wrote about a few months ago:
If you really want maximum personalization, a memory of past queries is not enough. You’ll want the AI to see your grocery list, your browsing history, your social media posts, your email, etc.
To have that, you can’t just be a model maker. You need to be one of the tech behemoths that already has one or more of those data sources. Because if model performance hinges on “who knows you better”, then the model-only companies are bringing a knife to a gun fight.
But app makers, the last piece of the puzzle, probably have mixed emotions…
Why?
Siri’s New Powers Threaten the Apps
We need an example! Ok ok let’s do it.
Let’s say you built an app called “Taybull” (nice name btw) for restaurant reservations. If you use Apple’s SDK to write APIs that “talk” to Siri, then your users get those 5 new Siri powers we talked about.
But if you look below the surface – not all of these are long-term “good” for Taybull.
What is good?
A full in-app menu (#1), multi-modal Siri (#2), and natural language understanding (#5) are all good. These make Siri better at doing things in the Taybull app. When users feel empowered and have magical experiences in your app, they’re more likely to keep using it.
What is bad?
Actions without opening (#3), and Spotlight Siri (#4) are not good. If you think about it, these are capabilities that reduce the likelihood of users seeing or truly depending on your app.
Apps as a set of static actions
If I can book reservations in Taybull without opening the app… I might stop noticing the beautiful interface you’ve designed. I might not even see your new messaging feature – designed for friends to hype-up meals and drool over menus together. Taybull is now confined to only its core actions.
Apps as commodities
Taking that even further… if I’m not spending any time in Taybull’s app itself, and I’m only using it to take actions, then I might even start to forget its name. If I can use Taybull, or FoodTime, or Dinr (all fake, go crazy!) to do the same things, I might use them interchangeably. I don’t have to worry about feeling too “spread-out” or decentralizing my restaurant bookings… Even if I have 100 restaurant apps, Siri can easily make new reservations or find existing ones.
Wrapping Up
The carrot – the reward for going all-in on the new Siri capabilities, is that app makers will get to enjoy the same thrill as Apple. They can help deliver magical experiences to users.
The stick – the punishment if you don’t, is that your competitors might. And if they do, they can steal your users, because we (the users) will absolutely be enticed by magical Siri-enabled experiences.
So because of the prisoner’s dilemma that sets up for app makers, I think we’ll see a bunch of them do it – building for Siri as quickly as possible.
In the process, we might even see apps that are #4 or #5 in a category flip the table on today’s winners…
Huge opportunity here!
Bonus Bullets
Quote of the Week
“People who go into entrepreneurship to get rich aren’t going to be happy. It’s the building of things that makes you happy. You have to enjoy the process whether you succeed or fail.”
– Catrina Fake, Co-Founder of Flickr
Quick News Reactions
SEC backs off of ETH – The SEC was investigating the #2 cryptocurrency to determine whether sales of ETH were illegitimate securities transactions. Apparently, they’re closing the investigation and not planning to bring any charges. This time last year, things were much bleaker (as we covered here), so it’s nice to see crypto scoring small wins where they can.
Netflix is trying to be Disney – They’ll have two physical locations – one in King of Prussia, PA, and one in Dallas, TX. And fans can immerse themselves in a rotating set of experiences that bring the Netflix stories and characters to life. I have to admit, this feels unnecessary and counter to the tech ethos of doing things that scale… I had similar criticisms of Airbnb’s in-person stunts a few weeks ago…
AI Drama Part 6000 – Ilya Sutskever (who helped found OpenAI and then left it in May), has started an OpenAI competitor to chase what he’s calling Safe Superintelligence. It’s crazy to see yet another influential early OpenAI backer breaking-off to start their own thing in response to perceived reckless or risky OpenAI leadership (Anthropic was started from the same point-of-view).