Open Editor's Digest for free
Rula Khalaf, editor of the Financial Times, picks her favorite stories in this weekly newsletter.
A decade ago, after selling Siri to Apple, the creators of the voice-controlled AI assistant had another big idea.
With their new company Viv, they set out to solve a persistent problem for smartphone users: how to accomplish everyday tasks without having to juggle multiple apps. Wouldn't it be great if you didn't have to switch between separate travel, hotel, and map apps when planning a vacation, and could instead rely on one that integrated them all?
Viv never made the progress it was looking for and was later integrated into Samsung's AI assistant Bixby. But in technology, good ideas rarely die: they just wait for advances in the underlying technology to make them possible. As with many other things, this progress has come in the form of large linguistic models.
Creating what is essentially a new layer of digital plumbing between apps and websites like this doesn't sound like the most exciting use of AI. But it could end up bringing about significant changes in how people use technology and changing the balance of power in the tech industry.
AI agents acting on behalf of their users are the trend of the moment. Giving them the ability to work across different apps, websites, and digital services can have far-reaching impacts.
For example, AI startup Anthropic recently demonstrated an AI system that operates a computer screen in the same way a person would. The demo, called “Use the Computer,” demonstrated technology that culls information from various sources to fill out an online form — the kind of low-level, routine task that occupies the days of many back-office workers.
The idea of using software to replicate something a worker does on his or her computer screen has been around in a variation of it for years. This system is known as robotic process automation, or RPA, and involves programming virtual “robots” to perform tasks spanning different applications. The natural language capabilities of generative AI have given a new lease of life to this idea. Anthropic's technology is designed to operate a computer just as a person would, although the software is not yet adept at doing common things on a computer screen such as scrolling.
For many office workers, such services that replace routine on-screen tasks could be the first real manifestation of generative AI. Matt Garman, head of Amazon Web Services, described his company's latest effort to automatically coordinate work between groups of AI agents so they can complete more complex tasks as “RPA on steroids.”
The closest thing to Viv's idea of integrating apps into the consumer world, meanwhile, has come from Apple. Known as App Intents, it requires developers to adapt their apps to work with Apple's AI, enabling software to run between apps without the user needing to open them.
The implications of this seemingly trite idea could be far-reaching. If an AI assistant or agent can automatically access any data or function you need to complete a given task, you will open fewer apps, visit fewer websites, and utilize fewer digital services. In effect, this will concentrate a person's digital activity in fewer places.
One result is likely to be a rush by app developers to ensure they remain one of the primary places that continue to control user attention and act as hubs for task completion, rather than risk seeing their apps relegated to sub-state status.
Ultimately, a group of multi-purpose AI assistants, acting like automated super-apps, may dominate. If so, it could suck value out of many standalone apps, as users no longer open them or shower them with interest.
This presents a dilemma for application developers. Opening up their services to AI for big tech companies could mean losing their direct relationships with users. But trying to stand out from one another could exclude them from the new digital ecosystems that are likely to cluster around AI agents.
For their part, big tech companies will be able to ensure seamless integration between their AI customers and their own applications, giving people more reasons to be drawn to their technology.
This will pose a new challenge for antitrust regulators. Just as they begin to internalize the way the largest tech platforms direct users to their internal services — a practice known as preferentialization — an entirely new layer of technology could emerge that connects tech users more tightly to the digital worlds of big tech companies. .
