Artificial Intelligence (AI)

Like plastic, AI is permeating every corner of life. But there is still time to manage AI better than we did with synthetics.

As the Grok case demonstrated, AI harms will continue to unfold without trusted channels between governments, industry and civil society.

Surveys how the 2025 ICRC GC IV Commentary integrates technology into its analysis of specific rules, while raising concerns about its treatment of data as property.

The 2026 REAIM Summit demonstrated the ongoing misalignment between global initiatives, industry narratives around AI capabilities, and the realities of military AI use.

The Trump administration’s salvo against Anthropic is invalidated by statutory limits, First Amendment freedoms, and the Constitution’s absolute bar against bills of attainder

Policymakers and the private sector should work with civil society to ensure that the pursuit of “sovereign AI” does not undermine fundamental rights.

Abuse of a tailored national security authority to resolve an ideological dispute playing out over DoD’s desire to change its contractual terms should not be taken lightly. 

AI companies should resist the urge to undercut Anthropic and use their collective leverage to make clear that access to frontier models comes with limits.

To lead in biotechnology, the United States must modernize its data infrastructure and create biodata built for the AI future.

The U.S. withdrawal from international cyber organizations will hamper intelligence sharing, coordinated response, and joint capabilities.

The Grok case suggests that effective AI regulation may come not from comprehensive AI-specific frameworks, but from applying existing harm-based laws to new capabilities.

In 2026, the nuclear order will become more fragmented, less predictable, and increasingly difficult to govern through existing institutions.