TechAILatest.newsTrackers
AI Business & Funding / Acquisitions

Palo Alto Networks agrees to buy Applied Compute for $4.5 billion in stock

Palo Alto Networks agreed on 18 November 2026 to buy Applied Compute, a startup run by former OpenAI researchers, for about $4.5 billion in stock.

By Emre A.Published

Abstract illustration of rising bars and a growth lineTech AI Latest
Illustration of rising bars and a growth line, made by Tech AI Latest for this story. Illustration: Tech AI Latest (Original illustration)

Palo Alto Networks agreed on 18 November 2026 to acquire Applied Compute, a startup founded in 2025 by three former OpenAI researchers, in an all-stock deal worth about $4.5 billion, subject to adjustments. Applied Compute adapts open-weight AI models to a company's own data and runs them in production. Palo Alto expects the deal to close in its fiscal third quarter of 2027 and announced it alongside quarterly revenue up 39% to $3.44 billion.

Key facts

  • Announced: 18 November 2026, with fiscal Q1 2027 results (Palo Alto Networks release)
  • Price: about $4.5 billion, paid in Palo Alto Networks stock, subject to adjustments
  • Closing: expected in fiscal Q3 2027, subject to regulatory clearance and customary conditions
  • People: Applied Compute CEO Yash Patil becomes co-chief AI officer after closing
  • Quarter: revenue $3.44 billion (up 39%), Next-Generation Security ARR $9.73 billion (up 66%)
  • Not stated in the announcement: Applied Compute's revenue or customer count

What Palo Alto Networks announced

Palo Alto Networks said on Wednesday, 18 November 2026, that it has agreed to acquire Applied Compute for total consideration of about $4.5 billion, subject to adjustments, payable in Palo Alto Networks shares. The announcement, filed with the SEC on the same day as the company's fiscal first quarter 2027 results, says the deal is expected to close in the fiscal third quarter of 2027, subject to regulatory clearance and other customary conditions. After closing, Applied Compute co-founder and chief executive Yash Patil becomes co-chief AI officer of Palo Alto Networks, and his team joins the company.

Applied Compute's platform lets a company adapt existing open-weight AI models to its own data and run them in production, the post-training and inference work that sits between a downloaded model and a working enterprise system. Palo Alto Networks says the purchase will let it train, deploy and secure open-weight models for its customers. Chairman and chief executive Nikesh Arora argued, in comments reported by CNBC and Yahoo Finance, that every enterprise will run a mix of open and closed models and that each of them will need to be secured.

Who Applied Compute is

Applied Compute was founded in 2025 by Yash Patil, Rhythm Garg and Linden Li, three former researchers at OpenAI. Benchmark led a $20 million first round in June 2025, according to Upstarts Media, and SiliconANGLE reported in October 2025 that the company had raised $80 million from Benchmark, Sequoia, Lux and angel investors. At that point it described its product as custom models and in-house agent workforces for enterprises, which it called Specific Intelligence, and named DoorDash, Cognition and Mercor as early customers.

Coverage of the deal by SecurityWeek and Yahoo Finance says the startup was valued at about $3.25 billion in a funding round earlier in November, which puts the purchase price nearly 40% above that mark only weeks later. Reports disagree on Applied Compute's revenue, and the announcement does not state it, so this story leaves the figure out.

The quarter behind the deal

The acquisition came with strong results. According to the earnings release, revenue rose 39% year over year to $3.44 billion, Next-Generation Security annual recurring revenue rose 66% to $9.73 billion, and remaining performance obligations grew 35% to $21.0 billion. GAAP operating income fell to $96 million from $309 million a year earlier, while non-GAAP operating income was $1.1 billion. The company raised its fiscal 2027 revenue outlook to $14.52 billion to $14.57 billion, according to CSO Online and Yahoo Finance.

On the earnings call, management said it expects Applied Compute to add about $1 billion of Next-Generation Security ARR by fiscal 2029, and warned of up to 100 basis points of operating margin pressure in fiscal 2027 and 200 basis points in fiscal 2028 before margins return to 29% to 30% in fiscal 2029. These are company targets, not results. CNBC reported that the shares rose about 6% on Thursday.

Why it matters and what to watch

Security vendors have spent the autumn buying AI companies, but mostly tools that watch or attack AI systems. CrowdStrike agreed in early November to buy Fiddler AI and Dreadnode for agent observability and offensive testing, and Cogent bought Specular for offensive security agents. Palo Alto Networks is buying something different: the capability to customise and host the models themselves. If it works, the security vendor becomes a supplier of the model stack as well as its guard, competing for the same enterprise budgets as cloud providers and model labs.

Three things are open. The share count is not fixed, so the final value depends on adjustments the release does not detail. Regulators still have to clear the deal. And Palo Alto has not said whether Applied Compute's existing customers, several of them AI companies, will stay on the platform once it belongs to a security vendor. More deals are tracked in the AI funding tracker and the AI business hub.

What changed

BeforeApplied Compute was an independent startup, reportedly valued at about $3.25 billion in early November 2026; Palo Alto Networks sold AI security through Prisma AIRS but had no announced product for adapting open-weight models
AfterPalo Alto Networks agreed to buy Applied Compute for about $4.5 billion in stock; Yash Patil becomes co-chief AI officer after a close expected in fiscal Q3 2027
Who it affectsEnterprises running open-weight models, Applied Compute customers, Palo Alto Networks shareholders and rival security vendors

Questions people ask

How much is Palo Alto Networks paying for Applied Compute?
About $4.5 billion, paid in Palo Alto Networks shares and subject to adjustments, according to the company's 18 November 2026 announcement.
When will the Applied Compute deal close?
Palo Alto Networks expects it to close in its fiscal third quarter of 2027, subject to regulatory clearance and other customary conditions.
What does Applied Compute do?
It adapts open-weight AI models to a company's own data and runs them in production. It was founded in 2025 by former OpenAI researchers Yash Patil, Rhythm Garg and Linden Li.

Sources

  1. Palo Alto Networks to acquire Applied Compute (press release, Form 8-K Exhibit 99.2) (Palo Alto Networks, 18 Nov 2026)
  2. Palo Alto Networks Reports Fiscal First Quarter 2027 Financial Results (Form 8-K Exhibit 99.1) (Palo Alto Networks, 18 Nov 2026)
  3. Palo Alto stock jumps on earnings beat and company's big bet on securing open-weight models (CNBC, 19 Nov 2026)
  4. Palo Alto Networks to Buy AI Startup Applied Compute for $4.5 Billion (SecurityWeek, 19 Nov 2026)
  5. Palo Alto Networks bets $4.5B on enterprise AI with Applied Compute acquisition (CSO Online, 19 Nov 2026)
  6. Palo Alto (PANW) Q1 2027 Earnings Call Transcript (The Motley Fool, 19 Nov 2026)
  7. Former OpenAI researchers launch Applied Compute with $80M in funding (SiliconANGLE, 30 Oct 2025)

Reference pages

Spotted an error? See the corrections policy and tell the editor. Facts in this story come from the sources listed above; vendor claims are attributed.

Comments

Loading comments.

No links, no insults; comments that break the rules are not published. 0/1000