Palantir CEO Alex Karp says AI’s “biggest problem” is a trust gap enterprises must close

Palantir CEO Alex Karp told CNBC that the central obstacle for the AI boom is not technical capability, but trust. Speaking in a recent television appearance on “Squawk Box,” Karp characterized the current moment as one where many businesses are excited by AI, yet still struggle to believe they can safely deploy it in real operations.

Karp’s comments were aimed at what he described as AI’s most painful bottleneck for mainstream use: the trust people need to justify using AI systems in high-stakes environments. In the interview framing reported by Yahoo Finance and TheStreet, he suggested that overcoming this trust gap may be more determinative than any specific advance in AI tooling.

The remarks come as enterprise buyers look for a practical path from AI pilots to production deployments, especially where errors, bias, and operational risk can carry meaningful consequences. Karp’s “trust crisis” framing places the emphasis on governance, accountability, and clarity around how AI outputs are generated and used, rather than simply on scaling adoption.

For Palantir, which markets itself to government and commercial customers with data-integration and decision-support systems, the positioning aligns with a broader enterprise sales message. The company’s pitch has long revolved around building AI-enabled decision workflows that can be audited and tied to real-world processes, an approach that can be easier to sell when a customer is worried about opaque automation.

Still, the CNBC interview account did not provide granular details such as specific customer stories, quantified adoption benchmarks, or timelines for when trust barriers are expected to fall. It also did not name particular regulators, standards, or product changes that Palantir plans to emphasize next, leaving open exactly which “trust” issues he meant most directly.

Market context matters because “trust” is now a cross-industry theme in AI. Companies across sectors are trying to manage concerns ranging from data provenance to model reliability, but the degree to which those efforts translate into concrete buyer decisions varies widely by industry and use case. Karp’s comments add Palantir’s chief executive voice to that wider narrative, positioning the company to benefit if enterprise demand increasingly rewards solutions that address governance and operational confidence.

As Palantir’s investors and customers watch for the next announcement on AI execution, the key question is whether Palantir will connect Karp’s “trust” framing to measurable outcomes, such as new deployments, contract expansions, or product enhancements. Another near-term watch item is whether the company’s management discusses any specific operational or regulatory steps it expects to matter most for enterprise adoption.

The one limitation is that the reported account centers on a television interview description rather than a detailed statement or transcript. Without additional disclosures tied to those remarks, investors will have to wait for Palantir to specify what “trust” means in practice, what problems it is solving, and how customers are responding.