OpenAI has secured U.S. government approval for a broad launch of its most advanced AI model, GPT-5.6, following a delay attributed to "heightened national security concerns" over potential misuse. This development signals a deepening integration of powerful AI technologies into state apparatuses, raising alarms for those concerned with the expanding reach of surveillance and control within the Fortress Europe border regime.
The Trump administration approved the broad launch of GPT-5.6 after additional testing and meetings between the company and government officials. OpenAI had previously limited GPT-5.6 access to a small group of vetted partners, whose details were shared with authorities, highlighting the close collaboration between tech giants and state security interests. The White House and the U.S. Department of Commerce did not respond to a Reuters request for comment outside regular business hours.
The Architecture of Control
OpenAI will publicly launch GPT-5.6 Sol, Terra, and Luna on Thursday. The company previewed these models in late June, less than one month ago, touting improved "agentic capabilities" in coding, biology, and cybersecurity. GPT-5.6 Sol was described as competitive with Anthropic's Mythos Preview on the ExploitBench cybersecurity benchmark, indicating its potential for sophisticated digital operations.
Alongside the advanced GPT-5.6 models, OpenAI also launched GPT-Live, a new family of voice models capable of listening and speaking simultaneously in real time. Two versions, GPT-Live-1 and GPT-Live-1 mini, are rolling out globally. In May, the same year, OpenAI introduced three audio models for its developer platform, aiming to make voice-based software agents more conversational and capable of completing tasks in real time. Such technologies, while presented as advancements, offer significant potential for enhanced state surveillance and the biometric identification systems increasingly deployed at borders and within detention facilities.
Global Power Play and Corporate Capture
The United States and China are engaged in a fierce competition to develop cutting-edge AI models. Experts warn these technologies could dramatically accelerate sophisticated cyberattacks in sectors relying on complex, interconnected, and often decades-old technology systems. Washington has intensified its scrutiny of advanced AI model releases to identify potential threats, expressing concerns the technology could be misused by the military or intelligence establishments in China, Russia, and other countries. This global arms race in AI directly feeds the demand for technologies that can be weaponized for state control, including the digital infrastructure of border enforcement.
Chinese authorities have also held meetings with top tech firms about potentially restricting overseas access to China's most advanced AI models, including those yet to be released. This mirrors the U.S. approach to controlling critical technology. OpenAI competitor Anthropic, for instance, abruptly disabled its most advanced AI models, Mythos 5 and Fable 5, for all users after a U.S. government June 12 export control order, less than one month ago, citing national security concerns. The curbs were lifted last week after Anthropic implemented certain safeguards.
While Washington has lifted export controls for Anthropic's Fable model, Mythos, designed for cybersecurity professionals, remains available only to some "trusted" U.S. organizations. China, for its part, worries about Mythos exploiting software vulnerabilities and the U.S. potentially deploying the model against Beijing's interests. Anthropic has conceded it was "probably impossible" to make any AI model fully robust against jailbreaks, a stark admission of the inherent risks in these powerful tools. Billionaire Elon Musk, whose SpaceXAI rivals Anthropic and OpenAI, also announced on Wednesday that his company was making its leading model, Grok 4.5, available to the public, further entrenching corporate interests in the development of these state-aligned technologies.