On theCUBE Pod: Nvidia steamrolls expectations as Mythos shakes up cybersecurity
SiliconANGLE Devony Hof ● Covered by 11 sources
Nvidia beat expectations again and is using the cash to deepen ties across AI and Cisco. Meanwhile, Anthropic’s Claude Mythos is jarring security teams enough to help CrowdStrike grow faster.
Based on reporting by SiliconANGLE, Devony Hof — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Nvidia keeps doing what Nvidia does: it printed another huge quarter, beat revenue expectations, and then immediately showed how central it’s become to the rest of the AI stack. The company was also tied to a reported $12.9 billion deal to buy Hugging Face, plus another investment in Perplexity. The money keeps moving, and so does the power.
On theCUBE Pod, John Furrier and Dave Vellante argued that the AI story has moved past the first rush of building GPUs and into something more operational. Vellante said the shift is from general-purpose enterprise IT toward accelerated computing, with Cisco now playing a bigger role because of its collaboration with Nvidia on rack-scale AI factories. Those systems pair Nvidia’s accelerated computing and Spectrum-X architecture with Cisco networking and enterprise operations.
The bigger point is that AI infrastructure is no longer just a cloud story. Furrier said the build-out is reaching enterprises, sovereign clouds and neoclouds, and even smaller installations like hospital data centers. The racks, in that view, won’t all be giant monsters anymore. Some will be much smaller and closer to the edge.
The software side hasn’t collapsed under that pressure, either. Salesforce beat expectations and its stock jumped almost 23% Thursday, which the hosts took as another strike against the idea of a looming “SaaSpocalypse.” Their argument is simple: bad software dies, but the stronger incumbents are being pushed to become AI native while also feeding the new infrastructure layer with data and workflows.
Then there’s security, where Anthropic’s Claude Mythos has clearly rattled people. The model’s release exposed how fast advanced AI can find and use weaknesses, and that helped CrowdStrike post $333 million in net new ARR, $45 million above the high end of guidance. Furrier called Mythos a “ChatGPT moment” for security. That sounds dramatic, but the underlying message is plain: one crack can now be enough, and agents make the blast radius a lot uglier.
My take — AI-written commentary, not fact-checked reporting
The market keeps treating AI like a software cycle, but this is looking more like an infrastructure takeover with a side of panic. Cisco getting “exceedingly relevant again” says plenty about who’s really winning here: the people who sell the pipes, the racks and the guardrails. Everyone else is just trying not to become a footnote in the billing system.
Read more about this at: SiliconANGLE
