Anthropic Suspends AI Models Following US Government Order on Foreign Access

2026-06-16

The BareStory

Artificial intelligence company Anthropic disabled its Fable 5 and Mythos models for all users on June 12. The suspension was implemented to comply with a United States government directive from the Trump administration that prohibits foreign nationals and entities from accessing the technology.

Although the federal order specifically targeted foreign usage, Anthropic disabled the systems across its entire customer base to ensure full compliance. The shutdown took place just days after the company launched Fable 5. On June 15, Anthropic representatives met with federal officials to discuss the restrictions.

According to the company, Fable 5 was developed by applying safety constraints to its Mythos model. Anthropic had previously claimed that the original Mythos system was too powerful to be released publicly due to its capacity to identify severe software vulnerabilities.

This intervention marks the second recent conflict between the company and the federal government. Earlier in the year, the Department of Defense labeled Anthropic a supply chain risk following a disagreement over the utilization of the company's artificial intelligence models.

Left Perspective

Overreach via Blanket Blackout
Weaponizing Supply Chain Designations
Gamble of Suppressed Transparency

Right Perspective

Shielding Strategic Cyber Assets
Forcing Total Compliance Architecture
Asserting Sovereign Tech Control

Left Perspective

• Overreach via Blanket Blackout Broad executive power fundamentally threatens open society when a targeted mandate results in the collateral suppression of public innovation. The Trump administration's directive prohibiting foreign access forced Anthropic to disable Fable 5 and Mythos for its entire global user base on June 12. Treating technological platforms as easily partitioned state assets ignores the interconnected reality of modern systems, effectively punishing all domestic civilian users for the actions of hypothetical foreign actors.

• Weaponizing Supply Chain Designations Institutional power structures frequently use bureaucratic labels to coerce compliance and stifle independent corporate policies. The Department of Defense previously labeling Anthropic a "supply chain risk" following a mere utilization disagreement suggests a pattern of state retaliation against private autonomy. This dynamic signals an alarming trend where the government wields its regulatory and procurement weight to punish tech companies that challenge the institutional status quo.

• Gamble of Suppressed Transparency Restricting access to vulnerability-detecting artificial intelligence fundamentally weakens broader societal tech resilience by consolidating knowledge within state silos. While Anthropic acknowledged the Mythos system was powerful enough to find severe software vulnerabilities, locking its safer iteration, Fable 5, behind a federal mandate prevents independent researchers from patching public systems. Shielding these models from the public sphere creates a centralized blind spot, prioritizing government control over decentralized digital safety for the vulnerable public.

Right Perspective

• Shielding Strategic Cyber Assets National sovereignty demands that advanced, dual-use technologies be strictly managed to prevent foreign adversaries from exploiting domestic infrastructure. Because the original Mythos system possessed the capacity to identify severe software vulnerabilities, releasing it without airtight borders functions as an uncontrolled strategic export. The Trump administration's directive prohibiting foreign access is a necessary application of the rule of law to protect national security interests from unauthorized offshore exploitation.

• Forcing Total Compliance Architecture Corporate prioritization of rapid deployment must never supersede the fundamental requirement of border security and regulatory adherence. Anthropic’s decision to completely disable the newly launched Fable 5 model on June 12 demonstrates that the company lacked the architectural safeguards necessary to isolate and block foreign nationals. Strict federal mandates correctly compel private entities to halt operations until they can guarantee absolute compliance, establishing a necessary precedent for civic duty over product launches.

• Asserting Sovereign Tech Control The state must maintain ultimate authority over systems that pose a fundamental risk to institutional continuity and national defense. The ongoing friction, evidenced by the Department of Defense previously designating Anthropic as a supply chain risk, highlights the systemic danger of tech developers defying federal security frameworks. Anthropic's June 15 meeting with federal officials reinforces the reality that private artificial intelligence developers must operate under the strict oversight of sovereign defense protocols, rather than acting as independent geopolitical actors.

How it may affect me

As a U.S. reader:

• In the short term, domestic civilian users and businesses have completely lost access to the Fable 5 and Mythos artificial intelligence models because Anthropic disabled the systems globally to comply with the federal directive.

• U.S. cybersecurity may experience mixed effects, as the restriction aims to shield domestic infrastructure from foreign adversaries but simultaneously prevents independent American researchers from using the models to identify and patch vulnerabilities in public software.

• Long-term access to future artificial intelligence tools may be delayed or restricted, as tech companies may be forced to halt operations or delay product launches until they can build strict verification architectures to successfully block foreign nationals.

• The ongoing conflicts and supply chain risk designations between tech companies and federal defense agencies indicate that future availability of advanced commercial technology will likely be heavily dictated by government security protocols rather than independent corporate timelines.

Anthropic Suspends AI Models Following US Government Order on Foreign Access

The BareStory

Left Perspective

Right Perspective

How it may affect me

Read the story at