Palo Alto Networks to Buy AI Startup Applied Compute for $4.5 Billion
The cybersecurity giant agreed to an all-stock deal for the open-weight AI model platform as quarterly revenue jumped 39% and it raised its outlook.
Palo Alto Networks, one of the world’s largest cybersecurity companies, said Wednesday it has agreed to acquire Applied Compute, a young artificial intelligence startup, in an all-stock deal valued at $4.5 billion. The announcement came alongside fiscal first-quarter results that beat Wall Street forecasts and a higher full-year outlook, sending the company’s shares up more than 7% after the report, CNBC reported.
What happened
According to a Palo Alto Networks press release filed with the U.S. Securities and Exchange Commission, the company signed a definitive agreement to buy Applied Compute, which runs a platform for post-training and inference of open-weight AI models. Open-weight models are AI systems whose underlying parameters are publicly released, allowing companies to download, customize and run them on their own infrastructure rather than relying solely on a provider such as OpenAI.
The release said Applied Compute lets businesses train, adapt and deploy open-source and open-weight models in public clouds or in so-called sovereign clouds, which keep data within a particular country or under a customer’s control. The $4.5 billion value is subject to adjustments, and the deal is expected to close in Palo Alto Networks’ fiscal third quarter of 2027, pending customary conditions including regulatory clearance.
Yash Patil, Applied Compute’s co-founder and chief executive, will become Palo Alto Networks’ co-chief AI officer once the deal closes, the company said. Applied Compute’s research team will join Palo Alto Networks to help embed AI across its products and train the company’s own proprietary models, and the startup will continue serving its existing customers.
Nikesh Arora, chairman and CEO of Palo Alto Networks, said in the release that Applied Compute “has built the best post-training and inference platform we’ve seen.” He added that the company’s Unit 42 threat intelligence group had reviewed the startup’s security foundations. Speaking to CNBC, Arora said the deal will help businesses secure open-weight models as their use grows, adding: “This could be bigger than what we do.”
A rapid rise for the startup
The purchase price represents a sharp step up from Applied Compute’s most recent private valuation. SiliconANGLE reported in early November that the company had raised $350 million at a $3.25 billion valuation in a round led by Elad Gil, a prominent angel investor, with participation from Nvidia, Shopify Ventures, HubSpot Ventures and more than a dozen others.
At the time, the startup said its annualized recurring revenue had topped $100 million less than 18 months after launch, according to SiliconANGLE. The outlet said Applied Compute was founded by Rhythm Garg, Yash Patil and Linden Li, and counts Microsoft and DoorDash among its customers. Its tools include reinforcement learning features for fine-tuning models and a router that directs requests to the most cost-efficient large language model.
Strong quarter and a higher outlook
The deal was paired with results that topped expectations. CNBC reported that Palo Alto Networks posted adjusted earnings of $1.05 per share for the quarter, compared with the 97 cents analysts surveyed by LSEG had expected. Revenue rose 39% from a year earlier to $3.44 billion, ahead of the $3.31 billion consensus estimate.
GAAP net income, which follows standard accounting rules, rose to $482 million, or 57 cents a share, from $334 million, or 47 cents, a year earlier. Next-generation security annual recurring revenue, a metric the company uses to track its newer cloud and AI-driven products, climbed 66% to $9.73 billion, according to CNBC.
The company raised its guidance for fiscal 2027. Key figures from the updated outlook, as reported by CNBC:
- Revenue of $14.52 billion to $14.57 billion, above the $14.16 billion analysts expected
- Adjusted earnings of $4.26 to $4.30 per share, compared with a $4.18 estimate
Why it matters
Palo Alto Networks has been steadily buying companies as it tries to position itself for demand for security tools tied to AI, CNBC noted. The company’s stock was up 85% for the year before Wednesday’s report. Arora said in the results release that AI is reshaping the threat landscape and that the company must keep evolving to protect customers.
The Applied Compute deal pushes the company further into AI infrastructure itself, betting that enterprises will increasingly run their own customized models and will need help building and securing them. Because the deal is paid in stock, Applied Compute’s backers will receive Palo Alto Networks shares rather than cash. The transaction still requires regulatory approval, and the company said it would share more details on its earnings call.


