The CHIPS and Science Act, along with the Biden administration’s export controls for semiconductors, have set the stage for a comprehensive industrial policy for advanced semiconductors. Now, a similar discussion is emerging around artificial intelligence, with federal officials eyeing a promote and protect approach to AI from a national security perspective. The work is still in its early stages, and it may take several years to fully develop a policy for AI.
An increase in government funding for domestic nondefense AI innovation would drive the promote aspect of the industrial policy. The Senate’s bipartisan working group has called for appropriations to reach at least $32 billion per year for AI innovation, as recommended by the National Security Commission on Artificial Intelligence. Senate Majority Leader Chuck Schumer and Senator Todd Young are leading the efforts in this area, but it may take time to turn the proposal into legislation and gain bipartisan support, especially given the high costs involved.
With little legislative action expected in Congress before the end of the year, opportunities for AI funding provisions to move forward are limited. The Senate working group is focused on building bipartisan consensus rather than rushing through partisan legislation. The House also has its own bipartisan AI task force that is expected to produce a report on addressing the technology, indicating potential overlap and differences between the two chambers in their approach to AI.
The second part of the industrial policy aims to protect advanced AI models from foreign adversaries, particularly China. The Commerce Department is considering export controls for the most advanced AI models, similar to those in place for semiconductors. Enforcing these controls may be more challenging given the lack of physical products involved, but the Biden administration is still considering moving forward with proposed regulations.
The criteria for determining which AI models would be subject to export controls have not been finalized, with potential factors including computing power, data used, and intended use of the models. While these efforts are still in the early stages, the focus on AI among officials suggests that this industrial policy could take shape over the next few years. The CHIPS and Science Act serves as a precedent for the development of policies in this area, indicating that discussions now will shape future regulatory priorities and potential obstacles in legislative efforts surrounding AI.
