AI Readiness Model for Operational Resilience<\/a> (ARMOR) as a strategic partner, contributing its Guardicore segmentation, API security, and Prolexic DDoS protection to a framework co-developed with NVIDIA. The technical bet is specific: offload security functions onto NVIDIA BlueField DPUs (dedicated chips that handle network and storage work independently of the main processor) so those functions stop competing with AI workloads for compute.<\/p>\nWhat this means for your business<\/h2>\n
The CISO who matters here is the one whose organization is already running or planning large-scale AI training clusters, not the one still doing POCs on a laptop. If your AI infrastructure is mature enough that security agent overhead is showing up as a real performance complaint from your data science or ML engineering teams, this architecture pattern is directly relevant. If you’re still consolidating your data estate, this is a future problem, and reading it as urgent is a vendor doing its job.<\/p>\n
The genuinely interesting claim in ARMOR is the DPU offload approach. Traditional endpoint security agents sit on the same compute plane as the workloads they protect, which means a GPU cluster running a billion-parameter model training job is also burning cycles on security telemetry. Moving segmentation enforcement onto BlueField DPUs creates what amounts to a separate enforcement plane, closer to how network hardware has always handled firewalling than how endpoint security has worked on servers. Akamai’s Guardicore, which it acquired in 2021, was already the strongest micro-segmentation play in the portfolio. Pushing it onto dedicated silicon is a credible evolution of that capability, not a marketing reframe.<\/p>\n
WWT positions ARMOR as vendor-agnostic, and Akamai, reporting as a participant in a framework it didn’t build, has an obvious incentive to make the partnership sound more complete than any single vendor’s coverage actually is. The honest read is that ARMOR gives large enterprises a structured way to audit AI security gaps across six domains, and Akamai fills some of those domains well. The question to pressure-test at renewal or procurement is which domains your current stack already covers and whether the DPU offload benefit requires a BlueField-based infrastructure investment you haven’t yet made. If your GPU clusters run on different hardware, the performance argument evaporates.<\/p>\n
Concept deep-dive: Microsegmentation<\/h2>\n
Microsegmentation is the practice of dividing a network into small, tightly controlled zones so that a breach in one area can’t freely spread to others, the way a fire door stops smoke from reaching every floor. In AI infrastructure, the risk it addresses is lateral movement: an attacker who compromises one node in a GPU cluster potentially reaching the data pipelines, model weights, or orchestration systems connected to it. ARMOR’s DPU offload moves this enforcement off the application layer and onto dedicated network hardware.<\/p>\n
Based on reporting from Akamai joins WWT\u2019s AI security framework to strengthen<\/a>, originally published 2026-07-09 08:33:00.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"Share with your CISO Akamai is betting that AI infrastructure security needs its own architecture layer, not bolt-on software. The company has joined WWT’s AI Readiness Model for Operational Resilience (ARMOR) as a strategic partner, contributing its Guardicore segmentation, API security, and Prolexic DDoS protection to a framework co-developed with NVIDIA. The technical bet is […]<\/p>\n","protected":false},"author":1,"featured_media":4943,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[143],"tags":[238],"tmauthors":[],"class_list":["post-4942","post","type-post","status-publish","format-standard","has-post-thumbnail","category-ai-security","tag-ciso"],"_links":{"self":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/4942","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/comments?post=4942"}],"version-history":[{"count":0,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/4942\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media\/4943"}],"wp:attachment":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media?parent=4942"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/categories?post=4942"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tags?post=4942"},{"taxonomy":"tmauthors","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tmauthors?post=4942"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}