SandboxAQ CEO Jack Hidary discussed security breaches at OpenAI and Hugging Face during a CNBC appearance on Wednesday [1].

These incidents highlight the growing vulnerability of the infrastructure supporting generative artificial intelligence. As major platforms face systemic risks, the stability of the global AI ecosystem depends on how quickly companies can secure their data and models.

Speaking on the "Squawk on the Street" program, Hidary said the recent security incidents involving OpenAI and Hugging Face underscore the critical need for robust security frameworks as AI integration scales across various industries [2].

Beyond immediate security concerns, Hidary commented on the sustainability of current AI spending. He said he weighed in on whether the massive investments pouring into the sector are yielding proportional returns or if the industry is facing a potential correction in capital allocation [3].

Hidary also analyzed the geopolitical dimensions of the technology race, specifically the competition between the U.S. and China. He said the AI race is a pivotal struggle for technological primacy, one that influences not only economic growth but also national security and global influence [2].

The discussion occurred on July 22, 2026 [1], providing a snapshot of the industry's current anxieties regarding both technical failures and international competition. Hidary's perspective as the leader of SandboxAQ, a company focusing on the intersection of quantum technology and AI, adds a layer of technical scrutiny to the conversation about how to protect these systems from sophisticated attacks [3].

“Jack Hidary addressed the recent security incidents involving OpenAI and Hugging Face”

The simultaneous occurrence of security breaches at leading AI hubs like OpenAI and Hugging Face suggests that the pace of AI deployment is currently outstripping the development of necessary security protocols. When coupled with the strategic rivalry between the U.S. and China, these vulnerabilities become national security liabilities rather than just corporate IT issues, potentially slowing the adoption of AI in sensitive government and infrastructure sectors.