Gone are the days of stumbling over your words to prompt Siri or Alexa to do your bidding: AI is making them quicker, smarter, and more like a prescient little human sitting in your pocket than ever before.
Advertised as having “life-bettering” reasoning, intelligence, and conversational skills, the assistant applicant pool has only grown in the last year: Would you like a digital employee that can summarize your work notes or help you study? One with a background in nutrition and GLP-1s that can give you dieting advice? (Reminder: They don’t have a medical degree.) Or maybe an assistant more like Andy in The Devil Wears Prada: One that can make your annoying shopping returns, send your more unsavory emails, or even grab you the latest popular book release before everyone else?
While developers, big-name tech companies, and even social media platforms have released a gamut of AI-powered tools — image generators, music makers, study aides — few are marketed as strongly to consumer markets as their digital assistants.
Tech giants have leaned heavily into digital assistants. Let’s look at what they are doing and how they view the future.
Chatbots walked, so next gen assistants could run right into our To-Do lists
Microsoft’s Copilot debuted last year as the go-to “companion” for Windows users, complete with the ability to move information across the Microsoft Suite, generate text and summaries, and a memory-like feature straight out of the Sci-Fi epics of our youth — the controversial Recall tool — can sift through your entire PC history to answer very specific personal browsing questions, like finding a blue dress you saw an ad for, Mashable’s Kim Gedeon reported.
Certain Copilot AI features are now built into Windows services.
OpenAI had announced a multimodal, voice-enabled version of its internet-scouring chatbot that same month, able to respond to conversation and requests more organically.
A few months prior, in a test run by WIRED, experimentally advanced AI voice helpers like vimGPT were able to conduct multi-step tasks across web pages, like subscribing to services and finding flight options.
“A year from now, I would expect the experience of using a computer to look very different,” Ishan Shah, vimGPT’s builder, said to WIRED at the time. “Most apps will require less clicking and more chatting, with agents becoming an integral part of browsing the web.”
Ruslan Salakhutdinov, former Apple head of AI, shared a similar viewpoint. “It will be so much more impactful if I ask Siri to do stuff, and it just goes and solves my problems for me,” Salakhutdinov said to WIRED.
In recent demos from Google and Apple, their assistants were able to process visual, auditory, and text information in real time, and respond back in more conversational ways than Siri and Alexa were ever able to. They can work across devices, apps, platforms — you name it — and utilize your “personal profile” to essentially predict the way you want a task to be completed, eliminating issues of access and making assistant interactions less frustratingly slow.
That was only four months after the theorizing of developers like Shah and researchers like Salakhutdinov.
And then Apple unveiled its vision for the future of Siri.
‘Assistants’ are the gateway to universal integration
At its annual WWDC keynote on June 10, Apple announced an internet-breaking Siri upgrade, making “the original intelligent assistant” more natural, more contextually relevant, and more personal, the company explained. Similar to Microsoft’s Recall, Siri can scour your devices to answer extremely specific requests.
By using on-device processing, Apple says it can find photos, summarize notes, and help you better understand Apple tools and hidden features without collecting your data. It can also do those pioneering generative AI tasks, like compose text-based stories and generate images. The company added that the initial update would be followed up with features making the OG even more “personal and capable.” It will even incorporate other AI models, too, like Google’s Gemini.
This “new era” for Siri is part of Apple’s larger “Intelligence” push, a less flashy, internally integrated, and all-around more subtle pitch for a next-gen assistant. Apple’s demure AI assistant showing underscores that the intelligence features aren’t exclusive to its productivity mascot, incorporating the tech across apps even when Siri isn’t in use. In fact, the company barely pitched the upgraded “assistant” at all, focusing instead on the ways it enhances tasks that Siri and Apple devices already strive to do.
Other companies are phasing out the term “assistant,” as well, opting for more generalized titles — and thus a more far reaching use case for its users — that hint at their future aims.
At Google’s annual I/O conference in May, CEO Sundar Pichai pitched the company’s new “AI Agents,” leveraging the term for the company’s bid in the AI-powered assistant market. AI agents, he explained, are “intelligent systems that show reasoning, planning, and memory” which will be able to “think multiple steps ahead.” Google’s agents aren’t contained to one app or use, either, able to operate across software and multiple platforms to streamline every aspect of your digital life.
“AI Agent” isn’t a term exclusive to Google. It’s frequently used to refer to AI tools that complete tasks and make decisions without “human-in-the-loop” roadblocks. This AI is doing things for you that you don’t even realize you need doing, until the option presents itself. Microsoft launched its own “Copilot Agent” in May, for instance.
The marketing of this term over “assistant,” appears to be a strategy in and of itself. Venture capital funds and investors are intensely interested in funneling cash into agents, over other generative AI uses.
But other terms are filtering in, as well. Earlier this month, work management platform Asana introduced its workflow optimizing “AI Teammates,” a title chosen specifically to “create a mental shift in terms of how people think of interacting with AI at work,” explained head of AI Paige Costello to TechCrunch. Teammates work with you, not for you, to reason and assign better work management decisions.
Design platform Canva avoids role-based titles all together, delineating its Gen AI assistant functions with the modifier “Magic.” Regardless of what they’re called, these assistants are being sold as helpers across industries, from teaching to business ownership to medicine, and not just for your home.
What’s next
Mustafa Suleyman, co-founder of Google’s DeepMind, said that AI assistants of the future will operate more like a personal “Chief of Staff.”
“It will be able to reason over your day, help you prioritize your time, help you invent, be much more creative … It will be a research assistant, but it will also be a coach and companion,” Suleyman told CNBC in an interview about the future of AI.
We are decades, if not centuries, away from the age of artificially intelligent robot butlers. As Mashable’s Mike Pearl writes: “Even though technology has accelerated to the point where we now have machines that can respond to simple written prompts with vibrant moving images of, say, fictional humanoid robots, or any other fantasy scenarios we care to conjure, physical robots only seem to bring joy to real-world humans if the human in question is named Jeff Bezos. Meanwhile, for average individuals, robots are mostly objects of frustration, if not outright fear.”
But we are living (and actively participating in) the age of faceless AI helpers. It’s technology that, unlike robots, is expected to only get better … and infinitely cheaper.
