Public Companies

Palo Alto bets $4.5B on secure open models

Palo Alto Networks will pay $4.5B in stock for Applied Compute to tune and serve open-weight models securely.

StartupHub.ai Staff

By StartupHub.ai Staff, reviewed by Daniel Singer, Editor

2 min read

Palo Alto bets $4.5B on secure open models
Image credit: StartupHub.ai

Why would Palo Alto Networks pay $4.5 billion in stock for a model-tuning shop?

It expects large customers to run a mix of open and closed models. Palo Alto Networks said that mix will need security around the open half, and Applied Compute is its bet to provide it.

Open-weight models are AI systems whose trained parameters are published so a company can download, adapt and run them on its own infrastructure. Post-training is the tuning stage that adapts that base system to private data, cost limits and safety rules before it answers real requests. Applied Compute sells a platform for that tuning plus inference, which is the live serving of models to users, across public cloud and sovereign setups that must keep data inside national borders.

That sovereign promise is the budget line to watch.

Platform and security teams now stitch together open models, hosting, guardrails and monitoring from separate vendors. Palo Alto Networks wants to collapse that work into one contract, Nikesh Arora, its chairman and CEO, said Applied Compute built "the best post-training and inference platform we've seen." The company said it formed that view as a prospective customer and after its Unit 42 Frontier AI Assessment checked the startup's security foundations.

Can a security vendor win training work? Palo Alto Networks argues trust is the wedge because no large provider now delivers tuned open models securely, according to the announcement. Yash Patil, co-founder and CEO of Applied Compute, said joining will put security at the center of its products, and he will become Co-Chief AI Officer after the close. The all-stock deal is valued at $4.5 billion, subject to adjustments, according to the announcement, and needs regulatory clearance before a planned close in fiscal third quarter of 2027. The news arrived with fiscal first quarter revenue up 39%, as Datatooza reported, a figure not in the acquisition release.

Watch the earnings call later today and the regulatory path to close for proof that security leaders will let one vendor tune and guard the models.

Sources

Get the free daily brief every morning

Every morning the day's AI news, new AI startups and new funding rounds. Free and takes a few minutes to read, unsubscribe at any time.

Join 64,000 other subscribers.

TopicsPalo Alto NetworksApplied ComputeYash PatilNikesh AroraUnit 42open weight AIPost-Training

StartupHub.ai StaffDaniel SingerWritten by StartupHub.ai Staff, Editorial teamReviewed by Daniel Singer, EditorAbout

StartupHub.ai Staff, Editorial team

The staff writers of StartupHub.ai, ranging from investment analysts to avid AI tool users, early adopters and critical enthusiasts. Backgrounds span engineering, business and the arts. We hold every piece to rigorous standards of research and review.

Daniel Singer, Editor

Daniel Singer is an AI expert who has been writing about, covering and building AI technology since 2018.

More from Daniel Singer

Test your Agent Readiness now

Paste your website. How well ChatGPT, Claude and the other agents can read and use a site, in about a minute.

Watch

The videos behind the news, one flip at a time

The feed mixes the videos from our articles with startup demos and you flip through them like TV channels for free with no signup.