The importance of the long-term future
We do research on how to do the most good in terms of reducing suffering. That is a difficult question to answer. All of our actions have many indirect and long-run consequences. However, we cannot simply ignore these consequences. If we not only care about our own generation but about future generations as well, these consequences matter. They will determine whether people lead flourishing or miserable lives, even if we don’t yet know how. That’s why our analysis must consider the outcomes of our decisions not just over the next few years, but all the way into the distant future. The latter effects could even be overwhelmingly important if we consider that there will be many more future generations and they’re likely to be more numerous than our own. We think there are good reasons to take this idea seriously. That’s why we’re focused on finding levers to steer human civilization away from outcomes that would cause unprecedented suffering—even if these scenarios lie far into the future.
The importance of artificial intelligence
We believe that smarter-than-human artificial intelligence (AI) could be such a lever. No other technology has similar potential to shape the future of our civilization in a significant and lasting way. It’s true that there is considerable disagreement when we will develop such technology and what exact shape it will take. But few experts question that it will be developed eventually and most of them agree that it would have profound implications for our society. Compared to anything else we could do, the influence of our actions on future AI systems could be particularly long-lasting because there are good reasons to believe that it will excel at self-preservation and goal-preservation. That’s why we believe that influencing the development and purpose of such a technology might predictably affect the future for thousands, perhaps even millions of years to come.
While there are also other ways of having a predictable impact on the long-term future of our civilization, we currently believe that work on making sure that advanced artificial intelligence is developed safely is particularly valuable. If you’re interested in learning more about our research, you can visit the website of our research project, the Foundational Research Institute, or get in touch with us.