Stay up to this point with free updates
Simply log in Artificial intelligence myFT Digest – delivered straight to your inbox.
A decade ago, after selling Siri to Apple, the developers of the voice-controlled AI assistant had one other big idea.
With their recent company, Viv, they wanted to resolve a persistent problem for smartphone users: tips on how to complete on a regular basis tasks without having to juggle multiple apps. Wouldn't or not it’s nice for those who didn't have to change between separate travel, hotel and map apps when planning your vacation and as an alternative could depend on one software that integrated every part?
Viv never made the hoped-for breakthrough and was later integrated into Samsung's AI assistant Bixby. But in technology, good ideas rarely die: they only wait for advances within the underlying technology to make them possible. As with many things, this progress got here in the shape of enormous language models.
Creating what is basically a brand new level of digital connection between apps and web sites hardly seems like the sexiest use of AI. But in the long run, it could lead on to vital changes in the way in which people use technology and a shift within the balance of power within the tech industry.
AI agents that act on behalf of their users are trending. Giving them the power to operate across different apps, web sites and digital services could have far-reaching implications.
AI start-up Anthropic, for instance, recently introduced an AI system that operates a pc screen like a human would. The demonstration, titled “Computer Usage,” showed how the technology collects information from various sources to fill out a web-based form – the form of routine, sure bet that occupies many back-office employees.
The idea of using a software program to copy exactly what a employee is doing on their computer screen has been around in one other form for years. So-called Robotic Process Automation (RPA) involves programming virtual “robots” to perform tasks that span different apps. The natural language capabilities of generative AI have breathed recent life into this concept. Anthropic's technology is designed to operate a pc the way in which a human would, although the software isn't yet superb at doing common things on a pc screen like scrolling.
For many office employees, services like these, which replace routine screen-based tasks, might be the primary real manifestation of generative AI. Matt Garman, head of Amazon Web Services, described his own company's recent efforts to mechanically coordinate work between groups of AI agents in order that they can handle more complex tasks as “RPA on steroids.”
The Viv idea of integrating apps into the buyer world is the closest thing to Apple's idea. Known as app intents, developers would must customize their apps to work with Apple's AI, allowing the software to work between apps without the user having to open them.
The implications of this seemingly prosaic idea might be far-reaching. If an AI assistant or agent could mechanically access all the info or functionality it’s essential complete a selected task, you’ll open fewer apps, visit fewer web sites, and use fewer digital services. In effect, this may concentrate an individual's digital activities in fewer places.
One result’s more likely to be that app developers rush to make sure they continue to be one in all the important thing places that proceed to regulate users' attention and act as hubs for completing tasks, moderately than risk having their apps within the be demoted to secondary status.
Ultimately, a handful of general-purpose AI assistants that function like automated superapps could gain the upper hand. If that is the case, it could impact the worth of many independent apps as users stop opening them or listening to them.
This presents app developers with a dilemma. Opening up their services to AI from big tech firms could mean losing their direct relationships with users. However, attempting to stand out could end in them being excluded from the brand new digital ecosystems which are more likely to coalesce around AI agents.
The biggest tech firms, in turn, will find a way to make sure smooth integration between their AI agents and their very own apps, giving people more reasons to care about their technology.
This presents the antitrust authorities with a brand new challenge. Just as they grapple with the way in which the most important tech platforms direct users to their internal service – a practice generally known as “preferencing” – a completely recent layer of technology could emerge, tying technology users much more closely into the digital universes of the Big Tech integrates.
