The artificial intelligence ecosystem has been shaken by recent events highlighting the growing complexity between technological innovation, geopolitical governance, and economic sustainability.
What happened
Last week, Anthropic, a leading developer of AI models, found itself at the center of an international controversy. The Trump administration ordered the company to block access to its most advanced models, Fable 5 and Mythos 5, for all foreign nationals, including its own employees and users within the US Anthropic got hit by export rules nobody understands. The decision followed a White House directive to revoke access for SK Telecom, a Korean telecom giant, due to alleged ties to China The Korean Telecom Giant at the Center of Anthropic’s Mythos Controversy. This forced Anthropic to temporarily disable access to the models for everyone, revealing the fragility of the global AI infrastructure in the face of political decisions.
Concurrently, the AI landscape saw other significant developments. Midjourney, known for image generation, unveiled its first hardware product: a full-body ultrasound scanner, marking an unexpected expansion into the medical sector Midjourney goes from generating cat images to full-body ultrasound scans. In Shenzhen, China, a new profession is emerging: operators controlling humanoid robots via VR systems, an innovation redefining human-machine interaction and the future of work Operating a Humanoid With Your Body Is a Hot Job in China’s Hardware Capital. However, not all news points to unlimited expansion. Several companies, including Uber and Meta, have reduced their use of generative AI due to high operational costs, in what has been termed an "ROI reckoning" NEA’s Tiffany Luck on AI IPOs, personal agents, and the ROI reckoning.
Why it matters
Anthropic's situation raises crucial questions about AI governance and access to advanced technologies. The decision to block access to critical models for national security or geopolitical reasons can have significant repercussions on research, innovation, and international collaboration. The idea that a company can be forced to disable its services globally to comply with specific regulations of a single country highlights the need for a clearer, shared international regulatory framework. This directly impacts companies' ability to operate and users' ability to benefit from innovations, challenging the principles of openness and accessibility that have characterized technological development.
In parallel, Midjourney's expansion into the medical sector and the emergence of new professions linked to humanoid robots demonstrate AI's transformative potential in vital sectors like healthcare and industry. However, the "ROI reckoning" underscores that AI adoption is not without economic hurdles. Companies must balance the desire for innovation with the need for financial sustainability, leading to a more pragmatic evaluation of the effectiveness and cost-benefit of AI solutions. This means that access to advanced AI is not just a matter of regulation, but also of economic capacity, potentially creating divides between those who can afford to implement and maintain these technologies and those who cannot.
The HDAI perspective
These recent developments reinforce the conviction that ethical AI and robust governance are more necessary than ever. The Anthropic case demonstrates how political decisions can have an immediate and profound impact on technology access, raising questions about transparency, fairness, and digital sovereignty. For Human Driven AI, it is fundamental that AI development and deployment are guided by principles that center humanity, ensuring benefits are widely distributed and risks are mitigated through global dialogue and clear regulation. This is not merely a technical problem, but a matter of balancing innovation, security, and fundamental rights. Topics such as data sovereignty, international collaboration, and the social and economic impact of AI will be central to discussions at the HDAI Summit 2026 in Pompeii, where we will seek to outline pathways for a more responsible and inclusive AI future.
What to watch
In the coming months, it will be crucial to observe how Anthropic and other AI companies adapt to this new landscape of export controls and how governments will seek to balance national security and innovation. It will also be interesting to monitor the evolution of generative AI costs and the strategies companies adopt to maximize ROI, as well as the impact of new AI applications in sectors like medicine and robotics on the labor market and society.