.avif)
In this Telescope LENS Q&A, we talk with Galen Hines-Pierce, founder and principal of the Minerva Fund and co-founder and vice chair of the Recoding America Fund. An independent researcher on artificial intelligence and its impact on human institutions, Galen shares insights on how societies adapt to technological transformation, why the real rate limiter for AI is public buy-in, and what it will take to build a new social contract between people and technology.
We are obviously on the precipice of yet another major technological change, perhaps the most important and transformative yet. But as we know from past periods of technological disruption, how societies adapt and diffuse these technologies — how they engage in what I call, in a paper coming out with some colleagues at Microsoft and DeepMind and other frontier model labs, the problem of “reverse alignment.” Not aligning the model to user intent, but how society aligns itself to new technologies to unlock broad-based prosperity and widespread benefits. That historically is the differentiator between a society that might have access to a technology early and then it kind of just sits stagnant for decades, potentially even centuries, versus another society that winds up capturing it and spreading its benefits widely in a way that reinforces prosperity, security, and broad social buy-in.
My worldview is directly informed by this idea that the rate limiter for artificial intelligence, or technology broadly, is buy-in. And to get that buy-in, you need to create more and better bets, shots on goal — attempts at ways that these technologies will benefit people and that those benefits will be legible early and often. Otherwise, the political economy winds up overwhelming technological progress, and that becomes the break in the rate limiter. To some degree you are sort of seeing this already.
My hope is that the work by myself and colleagues, and the work that Telescope is doing, is helping to point towards a new social contract between people and technology — one that I think has eroded over the last decade-plus. That is the task in front of us. I don't think there is any way to bypass or shortcut it. People are rational. In aggregate, there is a sense that not all of the gains of technology have been widely shared and that there have been some winner-take-all dynamics. Now is a chance to revisit some mistakes of the past and think about a new social contract, and tech that is broadly beneficial for all parts of society.
I absolutely wish there was more questioning around not just what we do not want, but what we do want. How do we engage in better sense-making as individuals, communities, and then as a collective, a nation? How do we come up with a positive, affirmative vision for the things we want?
The thing I am working on, and I know many others are too, is thinking about how you use artificial intelligence to take massive amounts of unstructured qualitative data and start to extract signal from that — but then have it be part of a collaborative loop. So you are not just using it for surveilling people, you are using it in dialogic processes, in ways that help to better inform democratic decision-making, that make trade-offs more legible for people, so that they can make better and more informed decisions that map onto their actual needs, desires, wants, hopes, dreams for their imagined better future.
What gives me the most hope is seeing the growing recognition that this is not just a nice to have — that the rate limiter is going to be buy-in, and the same with the diffusion of these technologies. The question comes down to: what are the lessons we have learned from the past? What rhymes with some of the changes we will see? What are the levers and tools available within market democracies — not to either overly retard progress or to exempt everything from small-d democratic legitimacy, but what is that middle path, that narrow corridor, that lets us capture as much of the upside while reducing harms and using these things safely. That is the work. And I believe we are up to it.
