Zero revivals. That’s the score for Centripetal Networks after today’s Federal Circuit smackdown.
In a world where network security patents sprout like weeds — only to get yanked by the PTAB — this one’s particularly brutal. The court affirmed the Board’s call: U.S. Patent 10,284,526? Unpatentable. Anticipated by a Cisco user guide from years back. And all because ‘allowing’ a packet to pass? Yeah, that’s an action, folks.
Look. Centripetal thought they had something clever. Decrypt packets, peek inside, do a ‘corresponding action,’ re-encrypt, ship ‘em off. Sounds secure, right? Except the PTAB — and now the CAFC — said nah. That action includes just letting the damn thing through.
‘Allowing’ Isn’t Inaction, It’s the Point
Centripetal whined. They claimed this reading makes the ‘transmit’ step redundant. Like, why bother re-encrypting and sending if ‘allowing’ already does it? But Judge Lourie wasn’t buying.
“Allowing a packet is the opposite of blocking it and would naturally be considered an action.” – CAFC
Boom. There it is. Straight from the opinion. The spec mentions blocking as an example — so allowing’s the flip side. Logical. No superfluous nonsense. The court nailed it: re-encrypt and transmit happen after the action. Distinct steps. Centripetal’s left holding an empty bag.
And here’s my hot take, one you won’t find in the dry legalese: this echoes the Theranos of patents. Flashy claims, but prior art’s been doing the basics forever. Cisco’s IronPort AsyncOS 7.1 User Guide (IPUG)? Publicly accessible, indexed enough to count. Centripetal griped about searchability — court said evidence says otherwise. Tough cookies.
Short para for emphasis: Patent trolls, take note.
The PTAB hit claim 1 on four fronts — anticipation by IPUG and Altman, plus obviousness combos. CAFC only needed one. IPUG anticipates. Done. Others? Moot. Efficient. Brutal.
IPUG’s magic? It decrypts traffic, applies policies: block, redirect, or allow. Allow? Then encrypt and send. Matches the claim beat for beat under the right construction. Substantial evidence? Piles of it.
Why Was IPUG ‘Printed Publication’ Prior Art?
Centripetal’s Hail Mary: IPUG wasn’t public. No Google juice, they said. Keysight showed the webpage was out there, accessible to the skilled. Court: supported by evidence. End of story.
But dig deeper — this isn’t just semantics. Network security’s patent thicket is already a mess. Remember the post-Alice bloodbath? Software patents claiming abstract ideas with tech sprinkles? Shredded. This fits right in. Centripetal’s trying to patent the equivalent of ‘check email, then forward if okay.’ Not novel. Not non-obvious.
Bold prediction: expect more cybersecurity firms to sweat their IP portfolios. If ‘allowing’ counts — and it does — what’s left? Pure math breakthroughs? Good luck.
Centripetal’s the poster child for overreach. They own the patent, sue Keysight, lose at PTAB, appeal, lose again. Costly lesson. Keysight? Smiling. Probably popping champagne in their labs.
Does This Doom Other Network Patents?
Not entirely. But it’s a warning shot. Claims need teeth — specific actions beyond pass/fail. Vague ‘perform corresponding action’? PTAB bait. And CAFC’s deferential to Board findings. Substantial evidence rules the day.
Historical parallel? Think Amazon’s 1-Click patent saga. Hyped, litigated, eventually diluted. Centripetal’s on that path. PR spin from patent holders will claim ‘nuanced,’ but courts see through it. This decision strips the hype.
Dry humor aside: if your patent’s novelty boils down to ‘or let it go,’ you’re cooked. Rewrite those claims, folks. Or don’t — save the lawyers the bill.
The ripple? IPRs keep humming. PTAB invalidates 70% of challenged claims these days (rough stat from recent years). Centripetal joins the pile.
One long wind: In an industry paranoid about breaches — where every packet’s a potential bomb — patents like this promised selective scrutiny without full decryption overhead. Noble. But if prior art already nails the flow, what’s the invention? Efficiency tweaks? Obvious. The court’s construction forces patentees to claim the how, not the what. Smart move. Prevents monopolies on basic flows.
Keysight’s win underscores: challengers with solid prior art references thrive. IPUG wasn’t some obscure memo — user guide for web security. Public. Actionable.
Why Does This Matter for Cybersecurity Firms?
Simple. Your moat might be illusory. Investors, perk up: portfolio audits incoming. Centripetal’s stock (if public) would’ve dipped — private, but partners feel it.
Critique the spin: Centripetal’s appeal brief probably oozed ‘innovation protector.’ Court: prior art wins. No sympathy.
Wrapping the tech angle — efficient decryption’s key in high-volume threats. But patenting the sequence? Risky post this.
🧬 Related Insights
- Read more: Why AI’s 50% Accuracy Problem Makes It Unfit for War—And Why Big Tech’s Safety Promises Don’t Matter
- Read more: Ex-FCC Chairs and EPIC to SCOTUS: Fine Carriers for Selling Your Location Tracks
Frequently Asked Questions
What happened in Centripetal Networks v. Keysight?
Federal Circuit affirmed PTAB: patent claims unpatentable as anticipated by prior art like Cisco’s IPUG. Key issue? ‘Corresponding action’ includes ‘allowing’ packets.
Is IPUG considered prior art after this decision?
Yes. Court ruled it publicly accessible pre-filing date, with substantial evidence of indexing and availability.
Will this affect other network security patents?
Likely. Vague action claims vulnerable in IPRs; forces sharper drafting to avoid anticipation traps.
