Despite the public repeatedly rejecting Meta’s ideas for digital assistants, Meta CEO Mark Zuckerberg remains dedicated to the concept, which, in his view will make the world a better place.
This week, Meta released its Muse app and chatbot, the company’s most advanced artificial intelligence-powered assistant tool yet. Muse enables people to name their bot, engage with it, assign it tasks that it can undertake in the background, and basically utilize it as an ever-present personal assistant.
Zuckerberg sees this as the next evolution of AI and a key step on the path toward personal superintelligence to everyone.
But it’s not the first time that Meta has released basically this exact same thing.
In 2015, Meta launched its M personal assistant, which, like Muse, was an AI-powered tool that could undertake tasks on behalf of users.
What tasks could M complete?
As explained by then Messenger chief Davis Marcus: “It can purchase items, get gifts delivered to your loved ones, book restaurants, travel arrangements, appointments and way more.”
Essentially, M was extremely similar to Meta’s new Muse project. AI tools have obviously advanced dramatically in the past decade, so Muse presumably has more capacity for understanding and undertaking tasks. But the same kind of option already existed, in some form, more than 10 years ago.
However, nobody used it. Despite Meta’s enthusiasm for bots, user interest was limited, and Meta couldn’t justify maintaining the M project for very long. It was officially shelved in January 2018.
Meta also released celebrity-themed chatbots on Messenger, in three different forms, none of which generated much interest. The company also launched Messenger bots in 2016.
Nobody paid any of these features much mind, despite Meta going on a big promotional push, and getting celebrities to endorse the utility of its automated bot offerings, similar to what it's doing with Muse today.
Yet, Meta keeps on pushing the same thing. But based on past experience, it’s hard to see Muse catching on in any significant way, and that could seriously dent Meta’s broader plans for its massive AI investment.
So why is Meta so obsessed with a concept that clearly isn’t interesting to users?
Maybe it comes down to science fiction-driven fantasies of an ever-present computer system that can assist with tasks. Zuckerberg developed a system that does just that for his own home, so he clearly has a personal interest in that kind of functionality.
But mostly, it seems as though the desire to build a digital assistant is driven by optimization, a push to maximize every moment and to improve every decision, in order to live a theoretically better life.
Yet, that approach doesn’t appear to be shared by a lot of other people.
Many people appear to be content to live and learn, and to undertake tasks like product research and shopping on their own. People like to engage in regular human interactions, which Zuckerberg seems to see as wasted time, while most view it as experiencing life.
Indeed, in an interview with Sources this week. Zuckerberg outlined his own use case for the new Muse AI chatbot: “For me, when I'm using my Muse Agent, I kind of want it to help me be a better father and a better husband, and show up better for my friends.”
That statement points to what may be a perceptual misalignment with regular users. Zuckerberg approaches life with an optimization mindset and a view toward getting more things done in less time, and maximizing his efficiency.
Which is not a universal desire.
Zuckerberg’s passion for optimization is a valuable trait for programming, and for business. But again, not everyone shares that mindset.
That’s why Meta’s latest attempt to get people to use digital assistants will likely fail once again. Because despite its vast troves of data on the human experience, Meta views that insight as just that, data, and not as a catalog of the various ways in which people are looking to connect with the wider world.
