Anthropic claims its latest large language model could be used by hackers to exploit vulnerable code bases.
It provided a handful of big enterprises with early access to help prevent widespread cybersecurity breaches.
This company made it on the roster, and it could be more important than ever in the age of LLM-based attacks.
Anthropic announced Claude Mythos, its most advanced large language model yet, earlier this month. Announced is the keyword. The broad release of the artificial intelligence (AI) model has been delayed due to security concerns. Anthropic claims Mythos is so good at coding that it can identify and exploit vulnerabilities in codebases that have existed undetected for years.
To prevent its product from causing too much harm, Anthropic established Project Glasswing, an initiative that invites a handful of top enterprises to use Mythos to find and patch vulnerabilities across operating systems, web browsers, and other critical software before it is more widely released.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
While many view the improved capabilities of large language models as a major threat to cybersecurity stocks, the opposite may prove true -- at least for some companies. And Project Glasswing may have just crowned a cybersecurity champion.
Image source: Getty Images.
Among the list of companies with early access to Claude Mythos is Palo Alto Networks (NASDAQ: PANW). Palo Alto's position as a leading cybersecurity vendor, offering solutions across network security, cloud security, and security operations, undoubtedly played a key role in the company winning a spot in Project Glasswing.
Importantly, Palo Alto's scale may prove an increasingly valuable competitive advantage as large language models continue to improve. That's especially true if AI labs come to it first before releasing models to the broader public. Palo Alto's scale means it can afford to spend money on tokens to use these models effectively and cover a broad range of attack surfaces for enterprises. That could leave smaller companies falling behind Palo Alto in capabilities, and create new opportunities for it to expand through acquisitions.
As large language models lower the barrier to creating cyberattacks and speed up the discovery of vulnerabilities, demand for cybersecurity solutions should increase. Not only that, but enterprises will likely look to consolidate their cybersecurity vendors as their needs expand to simplify billing and ensure their security needs are fully covered.
Palo Alto should meet that demand with its platformization strategy. It sells platform customers a complete solution for their network security, cloud security, or other security operations. As of the end of the company's second quarter, Palo Alto had reached 1,550 platformization customers, up 35% year over year. Net revenue retention for the group was 119%, as customers take more services each year.
It's worth noting Palo Alto Networks' stock isn't exactly cheap. It trades for 13 times sales expectations and 49 times forward earnings. However, it should see steady revenue growth over the years and strong margin expansion as it shifts to more software-based solutions and continues to grow revenue per customer. With its strong position in cybersecurity looking better than ever with the jump in LLM capabilities and its access to Project Glasswing, Palo Alto Networks should be able to sustain strong margins expansion for years to come, justifying its current share price.
Before you buy stock in Palo Alto Networks, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Palo Alto Networks wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $498,522!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,276,807!*
Now, it’s worth noting Stock Advisor’s total average return is 983% — a market-crushing outperformance compared to 200% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
See the 10 stocks »
*Stock Advisor returns as of April 26, 2026.
Adam Levy has no position in any of the stocks mentioned. The Motley Fool recommends Palo Alto Networks. The Motley Fool has a disclosure policy.