OPINION:
After several years of legislative progress, executive action and diplomacy, America’s high-stakes semiconductor reset is entering a new stage. With that important work in the rearview mirror, now is the time to ensure that these efforts do not go to waste and that the United States can reap the promised economic and security benefits.
Last year’s CHIPS and Science Act, for example, is rightly positioned as one of the Biden administration’s major accomplishments — but its rocky implementation shows just how difficult it is to regulate and move a highly complex, integrated and optimized global supply chain.
To ensure the success of America’s new industrial policy, the federal government must take a number of steps to refine the details, consider future threats, and mitigate countermoves from adversaries such as China and Russia.
It is worth recalling the stated goals of the Biden administration and Congress. At its core, the CHIPS Act is meant to counter decades of economic trends that have seen chip production offshored to Asia and rebuild U.S. manufacturing capacity, which now commands 12% of the global market, compared with 37% in 1990.
This laudable but massive goal will not be easily achieved, and recent commentary suggests that the $52 billion investment might not work at all and that changes and clarifications to the policy are needed.
One clear avenue for a revision is to reduce the red tape holding up the release of federal funding to the companies that need it. Facing industry criticism, the Department of Commerce has since provided important guidance on the application process.
Still, the Taiwanese chipmaking giant TSMC, which has a high-profile facility under construction in Arizona, recently said that it is communicating with Washington about its “guidance” for the law amid concerns that the application may expose confidential corporate strategy — yet another unnecessary roadblock hindering a critical piece of legislation. More work is clearly required.
This is just the tip of the iceberg. The Biden administration has also taken a series of actions designed to curb growing Chinese investment in semiconductors: it recently introduced its New Export Controls on Advanced Computing and Semiconductor Manufacturing Items to the People’s Republic of China. According to the announcement, the updates will restrict China’s ability to obtain advanced computing chips, develop and maintain supercomputers, and manufacture advanced semiconductors.
This, too, is a sensible policy goal, but the devil is in the details. And in a high-stakes game, policymakers should ensure that they are proactively considering every move an adversary like China might pursue to skirt these new restrictions.
One such move, currently unimpeded, is open-source chip design infrastructure that is not subject to the recent Bureau of Industry and Security export controls. This is not hypothetical: Right now, an open-source technology in part funded by the Defense Advanced Research Projects Agency called RISC-V a viable alternative to commercial Western technology.
China knows it. Beijing is increasingly investing in RISC-V, and Chinese companies plan to use it more frequently. Notably, given its open-source nature, chips made here in the U.S. on RISC-V technology could end up in the hands of our most significant adversary — potentially for use in advancing their own military capabilities.
Regulating open-source technology through export controls is easier said than done, but one should look no further than how OpenAI developed, voluntarily excluding countries of concern like China. Other open-source technologies should mirror this approach to protect both the significant benefits that come from the open-source model while also protecting U.S. and allied national security aims.
Unfortunately, the caution and forethought that OpenAI demonstrated is the exception, not the norm, and without other actors willingly doing this, the U.S. government must take action to protect its stated national security objectives. Because the current proposed rules have not yet been established, there is an actionable next step for BIS to ensure that any new rules related to semiconductor technology apply to open-source technology such as RISC-V.
This may be the necessary spark among U.S. companies and academic institutions to institute their own controls — promoting a shared responsibility in the public and private sectors.
Much more needs to be done on the domestic front to bolster chip production, and much more needs to be done on the international stage to curb new and emerging threats. A firm foundation is already in place. Now, the Biden administration and Congress need to take these steps to ensure broad goals are not undermined by faulty rollouts and costly oversights — and the good news is there is still time to get it right.
• Michael J. Bear is a senior executive leader and management consultant with over 30 years of experience in strategy, international operations, and turnaround management. He is an adjunct associate professor at New York University.
Please read our comment policy before commenting.