In this issue, we dig into the question of how AI affects expertise and skills: how AI affects mastering a complex field like coding, preserving human expertise in the age of AI, and a conversation with Matt Beane about his book, The Skill Code.
AI's impact on skill development: Balancing productivity gains with long-term expertise. Explore how AI changes apprenticeships, challenges traditional learning, and affects industries. Learn strategies to integrate AI while preserving crucial human skills and mentorship.
A conversation with UC Santa Barbara Assistant Professor Matt Beane about his book "The Skill Code: How to Save Human Ability in an Age of Intelligent Machines."
Resolving a fundamental incompatibility with AI in human decision-making
When we first started working with AI, we did a survey of what people did and didn’t want AI to do for them.
Two things jumped out at us. First, the more personal a task is, the more there’s a social aspect, the less people want an AI to help. This means that people are comfortable with an AI that performs tasks for them like getting information or finding ways to save time or money. But they don’t want an AI to do social things like looking after children or being a friend. Humans like performing tasks with an intrinsic social reward themselves.
The second thing that jumped out is that people want an AI to help them make decisions like helping them to be healthier but they don’t want AI to make decisions for them like managing their health. This difference - help me decide versus decide for me - is a big deal when you consider the role of technology in our lives.
Agency is at the core of the human experience. We prize our ability make decisions, to act, and to choose our futures—individually and collectively. The act of making choices brings meaning to our lives. We live authentically by making choices of our own, not following the choices of others. And through making choices, we understand the meaning of our existence and we create our future selves. Agency provides us the power to become the authors of our lives.
Humans want AI that works with us, understands our intent and is more than just a prediction machine. The philosopher Andy Clark says that humans are “loopy.” we don’t think in straight lines or steps, we think in loops. We play with ideas in different contexts and interact with others who build on our ideas. Once an idea is out of our head and into the mind of another, it takes on new forms. Different minds place different weight on the inputs to human judgment. When we share and receive feedback with others, or when we use tools outside our minds, we alter our perspective. Our internal models change. Engaging in a problem, actively attending, discussing and interacting is a generative process—we literally change our minds as we use resources which are outside our brain. The challenge is that AI is too frequently attempting to affect our choices, not support them.
Today, we humans too frequently are working for the machines. We teach the machines all about us so that the machines can get us to click on something or buy something that the machine owners want to make money from. That isn’t a mind for our minds. That isn’t AI enabling our agency to choose our future selves.
This points to a fundamental incompatibility between humans and machines. It’s simply not possible to use machines to make better decisions without this also altering our agency. What does it take to maintain agency while using machines to make better decisions? Control the separation. Develop an intuition for the machine and learn what it means to trust data. Cultivate a life of curiosity and choice. Don’t do this to be unpredictable, do this to rewild your attention. Bias to explore, even if it’s only in your own head. Decisions are actions in the world. Move. We store incredible amounts of information in the world. To learn we have to be out there in it.
More resources for these ideas:
Book: Surfing Uncertainty, Andy Clark.
Paper: The Extended Mind, Andy Clark and David Chalmers
Book: Mind in Motion, Barbara Tversky
Other things that caught our eye.
Researchers using AI to find out if whales have language. Hakai Magazine explores the Cetacean Translation Initiative, which is using artificial intelligence to decode whalespeak.
How do you know? Aeon article on what makes something worth believing.
Kara Swisher and Jeanette Winterson talking in Sway on Sex Bots, Religion and the Wild World of A.I. Kara says, “Let me just say, you pin a lot of hope on A.I. and a better future. I am of a belief that A.I. is us. Crap in, crap out. You’re more — crap in, better world out.”
Why can’t we teleport? Primer on the physics of teleportation in Wired. Tagged in “the future of travel.”
Fanta’s halloween costume. Go Ducks!
The Artificiality Weekend Briefing: About AI, Not Written by AI
