Pentagon Labels Anthropic a ‘Supply Chain Risk’ Amid Dispute Over Military Use of AI

The conflict with Anthropic over how flexible their technology can be used for military purposes makes the U.S. Department of Defense designating the AI company a supply chain risk in their ongoing dispute with the Defense Department as a result of the conflict with the Department of Defense regarding the potential use of the technology for military purposes. It signifies the current state of the relationship between DoD and the military industrial complex and Silicon Valley. It represents the conflict between the rapid advancement of technology, the moral and ethical constraints surrounding technology, and the oppressive focus of the defense establishment.

This designation came as a result of numerous negotiations between Anthropic and the DoD. The Defese Department has wanted the company to provide the level of functionality to military users of the AI model known as Claude. In this case, the Pentagon implemented the supply chain risk designation because Anthropic requested that their technology (to be used in military contracts) be incorporated into their policies governing acceptable use.

Understanding the Meaning Behind the Label "Supply Chain Risk"

A vendor can be deemed a supply chain security risk and, as a result, be barred from defense contracting opportunities. This risk designation means the vendor has a ‘supply chain risk’ designation, which, in this particular instance, means a government contractor is barred from incorporating Anthropic’s technology in any military-related projects.

This designation has clear and direct repercussions for defense contract beneficiaries. With regard to the use of Anthropic’s AI systems, particularly Claude, defense-related work is classified as a ‘defense use case’ and, as such, will be subject to the withdrawal of the AI systems.

The designation will highly likely have little meaningful and lasting effect on Anthropic’s overall business. For instance, it is accurate to state that the majority of business for the majority of nonprofit defense contractors will be unaffected.

The Predominant Concern is the Intersection of Artificial Intelligence and Military Use

Anthropic is at the center of the dispute because of their refusal to remove protections that prevent their AI systems from being used for certain controversial activities. Company policy prohibits mass domestic surveillance, unmanned systems, and other means of lethal autonomous engagement.

Company policy prohibits the use of autonomous weapons systems because of the established risk of unintended harm to combatants and non-combatants.

In contrast, the Pentagon believes that there are legitimate military purposes that AI could be used for and the military cannot be restricted from using AI for “legal” purposes such as [military] decision-making and operational control. They believe that embedded regulatory guardrails in [military] AI systems would hinder rather than enhance the military’s operational use of emerging technologies.

This divergence in views resulted in an impasse that led to the supply-chain risk designation.

Anthropic is in strong disagreement with the Pentagon’s decision and has indicated that it is willing to challenge the designation in court.

The company described the move as unprecedented, stating that supply chain risk labels have only been applied to foreign adversaries or entities deemed to pose a national security risk, not American technology companies that partner with the Pentagon.

Anthropic's executives say the company endorses the deployment of AI technologies for national defense. However, they insist that some ethical boundaries cannot be crossed.

The company’s executives have reiterated that there are two "red lines" that must be observed: the company’s technology must not be applied to surveillance of the population or to autonomous weapons.

Anthropic insists that the retention of these safeguards is essential to the responsible development of AI and is, to some degree, a reflection of the core principles of democracy.

A Broader Debate on Military AI

The controversy is indicative of the broader debate on the militarization of AI.

The potential for AI applications in defense and military operations has been the subject of extensive research and planning. AI is expected to be integral to defense operations. Governments are exploring the use of AI in several critical defense operations, including intelligence analysis, battlefield logistics, target acquisition, and defense of information technology systems.

In recent years, the Pentagon has worked more swiftly to implement AI across the military. Initiatives like Project Maven aim to integrate machine learning into the defense sector to assist in formulating strategies and analyzing large quantities of data.

Claude, one of the AI models at Anthropic, has already been incorporated into some classified and national security initiatives, thus illustrating the growing strategic importance of AI at the Pentagon.

The rapid pace of AI advancement, however, has brought ethical and security concerns. The absence of safeguards is likely to create highly problematic situations, especially when the involved AI systems may need to exercise lethal force.

Proponents of ethical AI argue that stronger governance is required to prevent companies from unrestrictedly applying AI technologies in ways that the public would consider unethical.

The Impact on Silicon Valley and the Pentagon’s Decision

The Pentagon’s decision will likely redefine the parameters of collaboration between technology companies and the U.S. government.

The relationship between Silicon Valley and the defense sector has always been uneasy. The government has been dependent on rapid innovations from the private sector, while the businesses in the private sector have been concerned about the uses of their technologies in the defense sector.

Partnerships with big tech companies can be limited and fragile. One proof of this is the protests against War Contracts.

From an AI ethics policy, the case with Anthropic demonstrates the widening gap of AI ethics policy and national security. If corporations impose more of such policies, the same concerns will recur in the future.

Next Steps

Currently, the supply-chain risk designation means that partners with the Pentagon can no longer use Anthropic’s technology for warfare.

Actually the company has defended this policy and will probably bring it to court. If so, it can establish the extent of the government’s control over AI availability.

Other AI companies are probably watching this closely as it pertains to their own dealings with the defense agencies.

The outcome of the dispute will determine the future of AI warfare, ethics in technology, and national security, in that order.

As more defense systems are centered around AI, the frequency of such discussions will increase.