Google and Microsoft have used their developer conferences as a platform to show off their generative AI tech, and now all eyes are on next week's Worldwide Developers Conference, where Apple Intelligence is expected to make its debut.
The Cupertino-based company is under a lot of pressure: Apple is lagging behind in the AI race, and it may feel it needs to go all out to impress fans and shareholders, but that shouldn't mean overpromising on features.
Reliability First
Apple makes some of the most popular devices in the world, and its AI features should make them even more useful. Many of the AI-powered features require going back to the cloud to get answers or input. But if Apple can perform some useful functions locally on the device, users might ditch cloud-based tools in favor of always-on AI. Voice memos and offline transcription in the Notes app might fit that need.
Apple is likely to roll out notifications, web page summaries, basic text generation and photo editing features, but many browsers, note-taking apps and photo-editing apps already have these features, and to stand out, Apple needs to make its implementation as smooth and seamless as possible.
Privacy First
Apple is likely to reinforce its privacy-first approach, so it may not give Siri or AI-powered features free reign over all apps. Bloomberg reports that AI features will only be available on the iPhone 15 Pro and iPads or Macs with M1 or later chips, and will be opt-in. If this is true, Apple is still cautious despite being late in the adoption curve of AI features, and doesn't want to get caught up in a user backlash.
The company recently came under fire for its iPad “Crush” ad, which showed creative tools being destroyed in a hydraulic press. This was seen as Apple undervaluing creators, their tools, and the effort it takes to make art by stuffing it into a slim capitalist package. AI is already unpopular among creatives, so Apple may not want to offend them again, so it's likely to take a non-controversial approach.
Siri Improvements
The biggest change expected is that Apple will improve Siri to better understand user queries and provide more accurate results. Currently, Siri cannot multitask; if you ask the assistant to set a 10-minute timer and a 5-minute timer, it will set a 15-minute timer instead. We may not need the help of generative AI to solve these problems, but improving Siri will at least require generative AI.
When Siri can't access apps as deeply as expected, Apple could bring in its AI assistant to make users' lives easier by allowing them to create complex Siri Shortcuts to accomplish multi-step tasks.
Rumors on the street are that Apple will announce a deal with OpenAI to bring AI capabilities to its entire operating system. It remains to be seen how much of Apple Intelligence will be built on that deal. Given the AI hallucination problem, Apple may not want to get directly involved in content-related AI features for now.
Many companies make big promises about their AI-powered features only to be disappointed by inaccurate or biased results. Companies like Google and OpenAI have had to abandon their AI features due to errors and copyright issues. That's why companies may not want to rely on LLMs (large language models) for content generation.