NSA Spending Billions on AI; Hospitals Face $1B in Extra Coding Costs

The National Security Agency (NSA) is expected to spend billions on AI model testing, reflecting a significant investment in advanced technologies. The Blue Cross Blue Shield Association reports that AI-assisted coding will add nearly $1 billion to U.S. healthcare costs over two years. This increase parallels the historical rise in spending associated with electronic health records, as noted by David Brailer, who claims, "AI will raise overall health care spending, just as electronic health records did before it."

McLaren Health Care has reported an additional revenue of $1 million per month due to AI software from SmarterDx, which specializes in AI-assisted coding. This software reportedly generates an average of nearly $12,000 in extra revenue per case without changing treatment protocols, as highlighted by Dave Mazurkiewicz.

The Peterson Health Technology Institute has analyzed these trends, while the AI Verification and Evaluation Research Institute is proposing a levy for AI testing to address the rising costs. In the AI sector, OpenAI plans to spend approximately $856 billion on computing power through 2030, while Anthropic anticipates generating 75 to 85 percent of its revenue from usage-based contracts, with a gross margin exceeding 80 percent on its API business. Anthropic is also preparing for an IPO in November 2026.

The Congressional Budget Office estimates that establishing an AI risk center will cost around $20 million per year, further illustrating the financial implications of integrating AI across various sectors. Caroline Pearson notes that disputes in healthcare are becoming more frequent, stating, "Both sides now go several rounds per dispute 'because it's cheap.'"