Investing.com -- Wolfe Research told clients in a note on Monday that investors are focusing on the wrong risk when it comes to Anthropic's leaked AI model, which will reportedly include significantly better cybersecurity capabilities than previous models.

Analyst Joshua Tilton argued that the model, reportedly named Mythos, “has the potential to ignite a machine speed cyberwar the likes of which we have never seen,” creating a setup that ultimately benefits CrowdStrike.

The market’s concern, according to Wolfe, is that Anthropic could erode the pricing power of existing security vendors.

But Tilton wrote that investors should instead worry that threat actors will gain “offensive cyber capabilities… that will now move at machine speed,” making today’s human-speed defensive products less effective.

Wolfe Research said this shift is more likely to increase security spending than shrink it.

“We don’t think anyone hears ‘war is coming’ and chooses to spend less on guns and ammo,” the report said, adding that the base case is “minimal negative impacts” to near-term spending, with the potential for customers to actually boost investment as they rush to reinforce defenses.

Against that backdrop, Wolfe upgraded CrowdStrike to Outperform with a $450 price target, citing its “brand, armory of tanks, F16s, and Black Hawks,” a reference to its Falcon platform and broad product suite.

Tilton said he has “increased conviction in CRWD's ability to accelerate ARR growth this year,” calling that a likely catalyst for outperformance.

Wolfe expects Anthropic’s formal announcement in early April but believes even a strong model release could “reinforce our view on the impending cyber arms race.”