OpenAI's public launch of its most advanced AI model, GPT-5.6, was delayed by U.S. government requests citing national security concerns, a framing frequently used to justify expanded state surveillance and the militarization of borders across Europe. This delay highlights the deepening entanglement of powerful AI technologies with state control apparatuses, posing significant risks to the rights and freedoms of migrants and marginalized communities.
SpaceXAI on Wednesday launched its Grok 4.5 AI model, presenting it as the company's most intelligent offering to date, specifically designed for coding and agentic tasks. The company stated Grok 4.5 was trained across tens of thousands of Nvidia GB300 graphics processing units, emphasizing meticulous data filtering, deduplication, and quality scoring. Such capabilities, while presented as technical advancements, could be readily adapted for processing vast datasets of biometric information or for identifying individuals deemed “undesirable” by border regimes.
The Surveillance State's New Tools
The lucrative enterprise AI tools market is a key driver for these developments. SpaceXAI announced last month its acquisition of Anysphere, the startup behind the popular AI coding agent Cursor, in an all-stock deal valued at $60 billion. This massive investment underscores the corporate profit motive behind technologies that could ultimately serve to strengthen Fortress Europe's digital infrastructure. Grok 4.5 is immediately available through SpaceXAI's AI coding agent, Grok Build, in Cursor and via the SpaceXAI console developer portal using an API key. Its availability within the European Union is expected by mid-July, bringing these advanced tools directly into the continent's operational sphere.
Elon Musk, in a post on X, described Grok 4.5 as an “Opus-class model, but faster, more token-efficient and lower cost.” The company priced Grok 4.5 at $2 per million input tokens and $6 per million output tokens, positioning it competitively against rivals like Anthropic's Claude Opus 4.8, priced at $5 and $25 respectively, and OpenAI's GPT-5.6 Luna, at $1 and $6. This race to cut costs and expand capabilities for enterprise customers, as noted by Reuters, fuels a wave of new systems that can be leveraged by state actors for border enforcement, predictive policing, and the criminalisation of movement.
Corporate Profits and State Power
Musk's AI startup xAI was acquired by SpaceX in February, with Musk stating in May that xAI would cease to exist as a separate company, becoming SpaceXAI. This consolidation of power within tech giants further centralizes the development of tools with profound societal implications. OpenAI will publicly launch its GPT-5.6 Sol, Terra, and Luna models on Thursday. The company had previously limited GPT-5.6 access to a small group of vetted partners, whose details were shared with authorities, demonstrating a direct collaboration between tech developers and state security apparatuses.
OpenAI claimed GPT-5.6 Sol was competitive with Anthropic's Mythos Preview on the ExploitBench cybersecurity benchmark. While framed as cybersecurity, such advancements can easily be repurposed for surveillance technologies, enhancing the capacity of the border regime to track, identify, and deport individuals. The intensifying competition among AI developers to improve model performance, cut costs, and expand capabilities for enterprise customers is driving this rapid development. This global trend, including the emergence of Chinese developers delivering capable models at a fraction of the cost, ensures a constant supply of increasingly sophisticated tools for the digital architecture of Fortress Europe. The focus on “national security concerns” by governments serves as a convenient pretext for the deployment of these technologies, often at the expense of human rights and migrant solidarity.