Future of Life Institute

In the AI Story

FLI's strategy combines research funding, policy advocacy, and public communication. The research arm has supported foundational work on technical alignment, interpretability, and governance. The advocacy arm has produced high-profile letters and statements that translate technical concerns into positions legible to policymakers and the public. The communication arm maintains the most widely-consulted public resource on AI safety issues.

The Institute's trajectory tracks the evolution of Tegmark's own thinking about the wisdom race. Early work focused on research funding and aspirational guidelines. Mid-period work—culminating in the Pause Letter—embraced public advocacy of specific constraints. Recent work has shifted toward regulatory engagement, partnering with governments developing AI governance frameworks, including significant influence on the EU AI Act and UK AI Safety Institute.

FLI's existence represents one answer to the structural problem Tegmark's analysis identifies: that no existing institution is designed to think on the timescales the AI transition demands. Political systems optimize on electoral cycles, corporations on quarterly earnings, academic institutions on publication timelines. FLI's explicit mandate is to think across generations—and potentially across cosmic timescales—about the trajectory of intelligence and its consequences for the cosmic endowment.

The Institute has been criticized from multiple directions. Some argue it overstates existential risks to the neglect of near-term harms; others argue it understates risks by remaining engaged with the industry it seeks to constrain. Tegmark's response has been that navigating between overstatement and understatement requires continuous calibration, and that the only alternative—silence—would constitute abdication.

Origin

FLI was founded in Boston in 2014 by a group of researchers concerned that transformative technologies were advancing faster than institutional capacity to govern them. The founding vision drew on earlier existential risk work—Nick Bostrom's Future of Humanity Institute at Oxford, Machine Intelligence Research Institute in Berkeley—but added a specific commitment to public engagement and policy advocacy that distinguished FLI's approach.

Key Ideas

Research, advocacy, communication. Three-pronged strategy combining technical research funding with public-facing policy work.

Asilomar conference (2017). Produced the twenty-three AI Principles endorsed by over a thousand researchers.

Pause Letter (2023). Transformed the AI safety conversation into mainstream public discourse.

Statement on Superintelligence (2025). Advocated conditional prohibition, marking escalation from guidelines to hard governance.

Institutional longevity. Designed to operate on timescales existing institutions cannot sustain.