Nvidia, Cisco, and CrowdStrike are each building their own AI safety playbooks, a move that could slow the push for unified industry standards. The three companies have not coordinated their efforts, raising questions about how their separate frameworks will mesh.
Three companies, three approaches
The playbooks are being developed internally, with each company tailoring its guidelines to its own products and services. Nvidia, known for its AI chips, is focusing on hardware-level safety. Cisco, a networking giant, is addressing AI risks in data traffic and infrastructure. CrowdStrike, a cybersecurity firm, is concentrating on threat detection and response for AI systems. None of the companies have disclosed full details of their playbooks, but the lack of a shared blueprint is already drawing attention.
Why fragmentation matters
Without a common standard, customers who use products from multiple vendors could face conflicting safety requirements. A company deploying Nvidia hardware alongside Cisco networking and CrowdStrike security might have to juggle three different sets of rules. That complexity could lead to gaps in coverage or slow down adoption of AI safety measures. The fragmented approach also makes it harder for regulators to assess compliance across the industry.
Impact on unified standards
Industry-wide safety standards for AI have been a goal for years, but progress has been slow. The separate playbooks from three major tech players could further delay that effort. Each company has its own timeline and priorities, and there is no indication they plan to align their work. Without a single reference point, other companies may also develop their own playbooks, creating a patchwork of guidelines.
No timeline for collaboration has been announced. The companies have not responded to requests for comment on whether they would consider merging their playbooks or adopting a common framework. For now, the AI safety landscape remains a collection of individual efforts, with no clear path to a unified standard.
