Should I set up a personal AI agent to help with my daily tasks?

—Seek Help

As a general rule, I think relying on any kind of automation in your daily life is dangerous if taken to excess and can be off-putting even if used in moderation, especially in terms of personal interactions. An I have an agent that organizes my to-do list and gathers online links for further reading? Amazing. An AI agent that automatically messages my parents every week with quick life updates? Terrible.

The strongest argument for not incorporating more generative AI tools into your daily routine, however, remains that the environmental impact these models persist during training and generate output. With all that in mind, I dig WIRED’s archivepublished during the glorious dawn of this chaos we call the internet, to find more historical context for your question. After searching a bit, I came back convinced that you are probably already using AI agents on a daily basis.

The idea of ​​AI agents, or God forbid “agent AI,” is the current buzzword du jour for every technology leader trying to hype their new investments. But the concept of a automated assistant being dedicated to completing software tasks is far from a new idea. Much of the discourse around “software agents” in the 1990s mirrors the current conversation in Silicon Valley, where leaders of tech companies are now promising a future flood of generative AI-powered agents that trained to do online work for us.

“One problem I see is that people ask who is responsible for an agent’s actions,” read a WIRED interview with MIT professor Pattie Maes, originally published in 1995. “Especially things like agents spending too much time on a machine or buying something you don’t want for you. Agents will raise many interesting issues, but I am convinced that we cannot survive without them. “

I called Maes in early January to hear how his view of AI agents has changed over the years. He’s as optimistic as ever about the potential for personal automation, but he’s convinced that “too naive” engineers aren’t spending enough time solving the complexities of human-computer interactions. In fact, he says, their indifference could prompt another AI winter.

“The way these systems are built, now, they’re optimized from a technical point of view, an engineering point of view,” he said. “But, they’re not really optimized for human design issues.” He pointed out how AI agents still are easily deceived or resort to biased assumptions, despite improvements in the underlying models. And a misplaced trust leads users to rely on answers generated by AI tools when they shouldn’t.

To better understand the other potential pitfalls for personal AI agents, let’s break down the awkward term into two distinct categories: those that feed you and those that represent you.

Feeding agents are algorithms with data about your habits and tastes that search through a wealth of information to find what suits you. Sounds familiar, right? Whatever social media A recommendation engine that fills a timeline with tailored posts or a non-stop ad tracker that shows me those mushroom gummies for the thousandth time in Instagram can be considered as a personal AI agent. As another example from an interview in the ’90s, Maes mentions a newsgathering agent who was groomed to return articles he wanted. That looks like my Google News landing page.