**[David Sacks: True Enterprise AI Security is About "Control," Not Abstract Alignment Research]** Golden Finance reported on July 3 that David Sacks commented on the X platform regarding an interview with Palantir CEO Alex Karp. He noted that some traditional media interpreted Karp's remarks as "emotional expression," but in reality, his views highlight the core issue of enterprise-level AI security. Sacks stated that true enterprise AI security is not about abstract "AI alignment" research or government-style regulatory frameworks but rather about enterprises having full control over their own data, model weights, and computational infrastructure. This ensures that core intellectual assets are not absorbed by model providers and turned into competitive advantages for their own products. He cited Karp's perspective, emphasizing that enterprise clients are genuinely concerned about control over computing resources, models, and data stacks—essentially ensuring that the "ownership of production means" is not transferred. Sacks also referenced the controversy surrounding the collaboration between Figma and Anthropic. According to media reports, Anthropic's launch of Claude Design "caught its partner off guard" and was accused of entering the application layer domain of its ecosystem partners during product expansion, thereby altering the value capture structure. He further pointed out that similar patterns have emerged in the expansion of product lines like Claude Code and Claude Legal, where model capabilities extend upward into vertical application domains. Sacks argued that this trend indicates model providers are shifting from being "foundational model suppliers" to "vertical application competitors," increasing the risk of supplier dependency for enterprise clients. The essence of enterprise AI security, he concluded, is not about trusting the long-term promises of model providers but ensuring optionality and control at the model layer to safeguard their own data and commercial "alpha."