“`html
“Apple does things practically.” Or, “Apple is late because it’s perfecting the tech.” “Would you prefer being the first or the best?” These are just some of the recurring arguments you will find in any heated Reddit thread or social media post hunting for some rage bait clout.
Yet, there’s some truth to it, as well. And a whole lot of hidden tech that sometimes takes a decade to come out. Apple Intelligence is the best example of one such leap, and it’s being seen as Apple’s answer to the generative AI rush.
Apple Intelligence soft-launched late last month with the iOS 18.1 developer beta, giving people their first chance to try it for themselves. And that’s exactly what I did.
AI isn’t anything new for Apple
But first, a quick intro on how we got here. Remember the Apple Newton MessagePad? Its widely ridiculed handwriting recognition feature, internally codenamed Rosetta, shipped in 1995. Apple engineer Steve Salomon conceived the tech in 1987.
Its foundations were based on neural networks, with assistance from segmentation and a language model that ingested dictionaries and probabilistic grammar. Sound familiar? Well, that’s your typical AI sauce spilling everywhere on social media by AI hype guys.
By 2014, Apple had already moved Siri — without any public announcement — to the fruits of machine learning, such as deep neural networks, natural language processing, and convolutional neural networks.
The point I am trying to make here is that even the most mundane Apple products have been serving cutting-edge AI for years. From turning your nicest pictures into a memory film to learning your phone usage habits to extend battery life, machine learning has been at the iPhone’s heart for a while now.
Apple Intelligence was just an attempt to quell shareholders who were apparently unhappy seeing all those “ChatGPT can change your life and make you a millionaire” posts on X and concerned that their blue chip horse was lagging. It’s, therefore, no surprise that Apple Intelligence didn’t really elicit a wow from the WWDC 2024 audience. It took the safe side of practicality — “AI for the rest of us,” as Apple puts it.
Apple Intelligence has several different sides. It can summarize websites, intelligently sort your notifications, generate images, and more. It also promises to make you a better writer with Apple’s new Writing Tools — a suite of features that, according to Apple, “help you find just the right words virtually everywhere you write.” It’s one of the biggest focuses of Apple Intelligence, and it’s what I’ve focused on the most during my time with Apple Intelligence so far. Here are my first impressions.
Hands-on with Apple’s Writing Tools
“Everywhere words matter,” says Apple’s tagline for its Writing Tools feature. Is it any good?
Well, it works, but I won’t entrust it with my professional work. There are a couple of reasons for that. One, it tends to miss the point. Two, there’s a whole universe of ethical dilemmas associated with it. Just take a look at the email I wrote to an academic asking for her expert insights.
Apple Intelligence did a grammatically perfect job at rewriting it, but it missed the two most important and specific points I hoped to discuss with my source. For comparison, here is what I got from Claude with a basic “rewrite this” prompt.
“`
