- with readers working within the Automotive industries
DivX had a patent on streaming partially encrypted video. Netflix was using the technology. The licensing revenue was real.
Then the Federal Circuit looked at one phrase in the patent claim.
Not the technology. Not the innovation. Just where a modifier pointed in a sentence.
The phrase was ambiguous. The court applied proper claim construction rules to resolve it. Under the adopted construction, the prior art taught the claim. DivX lost the patent.
This wasn’t a story about bad technology or a weak invention. This was a drafting problem that became a business catastrophe.
The Pattern: Small Errors, Silent Until Catastrophic
Recent Federal Circuit cases tell the same story from different angles.
In Enviro Tech, the word “about” killed a patent because the specification didn’t define it clearly enough. The court said “about 7.6 to about 10” failed to inform skilled artisans about the scope with reasonable certainty.
In Netflix v. DivX, a single grammatical ambiguity about where a phrase pointed, resolved through standard claim construction rules, led to the Federal Circuit invalidating the patent over prior art.
Different words. Same devastation.
The through-line is this: patents look well-written on the surface but contain strategic errors that don’t reveal themselves until litigation. By then, you’re spending millions to enforce a patent that can’t survive claim construction.
What Actually Went Wrong in Netflix v. DivX
The claim included this phrase: “locating encryption information that identifies encrypted portions of frames of video within the requested portions of the selected stream of protected video.”
The fight was over what “within” modified.
Did it modify “encrypted portions of frames of video” (the nearest term)? Or did it modify “encryption information” (the farther term)?
The Federal Circuit applied the nearest-reasonable-referent canon, a standard rule of English grammar: absent a comma or other textual signal, a modifier attaches to the nearest term it can plausibly modify. Under that rule, “within” modified the nearest term, not the farthest.
Under that construction, the prior art taught the claim. The patent was invalidated.
Here’s what makes this case remarkable: the PTAB reached opposite conclusions on the same claim language — twice.
First, the PTAB adopted Netflix’s construction and still found the patent wasn’t obvious. Netflix appealed. The Federal Circuit reversed on the obviousness finding.
On remand, the PTAB adopted DivX’s construction and again found the patent wasn’t obvious. Netflix appealed again.
This time, the Federal Circuit said the claim construction itself was wrong. Under the correct construction, the prior art taught the claim. Patent invalidated.
Same patent. Same prior art. The outcome turned entirely on which construction you applied.
Why Seasoned Patent Judges Reached Opposite Conclusions
Claim construction isn’t easy.
You start with the ordinary meaning of the words. You check the specification to see if it’s consistent. You review the prosecution history. But before all of that, you have to apply the basic precepts of English grammar.
The rule: a modifier attaches to the nearest term unless there’s a comma or other textual signal pointing elsewhere.
The PTAB initially applied a different reading. The Federal Circuit enforced the grammatical default.
If experienced patent judges at the PTAB can reach opposite conclusions on the same phrase, what does that tell you about the vulnerability in your own patent claims?
It tells you this: claim construction is not always predictable. You don’t have a crystal ball. But you can be diligent about the things that create ambiguity in the first place.
The Real Cost of Patent Drafting Errors
In the U.S., patent litigation costs can reach over $10 millions for cases with significant stakes. Many of those disputes arise from unclear claims or inadequate descriptions.
The errors don’t show themselves early.
They’re invisible at filing. They’re invisible during prosecution. They only reveal themselves when enforcement is already expensive.
By then, fixing them is impossible.
What Founders Miss: The Claims Section Isn’t Fine Print
Founders treat the claims section like fine print. They trust the attorney and move on.
That’s the vulnerability.
The invention is defined by the claim. Not by what you built. Not by what you intended. By the words on the page.
Every word can and will be used against you.
In DivX’s case, the phrase that created the ambiguity was probably drafted without a second thought. It looked fine. It read fine. It passed through the patent office without issue.
Then Netflix’s lawyers found the opening — and the Federal Circuit’s claim construction rules closed the door on DivX.
How to Catch This Before It Gets Filed
You need to be involved at multiple stages of claim drafting, not just sign-off at the end.
Here’s the practical workflow:
Stage 1: Before application is submitted
Review the draft claims with your attorney. Make sure they cover the core aspects of your technology. Confirm the language matches what you actually built.
Look at every claim term carefully, particularly in the independent claims. If you see modifiers, ask what they’re modifying. If you see vague terms like “about” or “substantially,” ask how they’re defined in the specification.
Stage 2: During prosecution
If the examiner proposes claim amendments, review them with the same scrutiny. Amendments made under pressure to overcome prior art often introduce new ambiguities.
This isn’t about becoming a patent attorney. This is about understanding that the claims section is where your patent lives or dies.
The Dependent Claim Safety Net
In Netflix v. DivX, the patent owner could have reduced the risk with better claim architecture.
If a modifier can plausibly point to either of two terms, capture both scenarios in separate claims. Use the independent claim for the broader interpretation. Use dependent claims to narrow each alternative explicitly.
That way, if one interpretation gets knocked out by prior art, the other survives.
In this case, the prior art didn’t teach one of the possible implementations. A dependent claim structured around that implementation would have survived the obviousness challenge.
You’re not just drafting claims. You’re building a safety net.
The AI Opportunity: Catching Landmines Before They’re Buried
We’re living in the age of AI, and patent drafting is starting to catch up.
You can now feed cases like Netflix v. DivX and Enviro Tech into AI tools and ask them to assess your claims for similar issues. The AI won’t replace the attorney’s judgment, but it can catch the mechanical errors that human review misses.
Think of it as a grammar checker trained on decades of Federal Circuit decisions.
This frees the attorney to focus on strategy: Does the claim capture the core of the invention? Does it cover competitive implementations? Does it leave room to design around?
The goal isn’t perfect claims. The goal is claims that don’t fail on technicalities when the stakes are highest.
What This Means for Your Patent Strategy
Patents are valuable business assets. They protect market share and create licensing revenue.
But a patent is only as strong as its weakest phrase.
DivX learned this the hard way. A drafting ambiguity, resolved against them through standard claim construction rules, cost them a licensing deal with Netflix and years of litigation.
The error wasn’t visible when the patent was filed. It wasn’t visible when it was granted. It only became visible when enforcement was already expensive.
That’s the pattern: small errors, silent until catastrophic.
You can’t eliminate the risk entirely. Claim construction will always involve judgment calls. But you can reduce the vulnerability by being involved in the drafting process, reviewing claims at multiple stages, and using tools that catch ambiguities early.
The alternative is discovering the problem when a federal judge is reading your claims in open court.
By then, it’s too late.
About the Author
Babak Akhlaghi is a registered patent attorney and the Managing Director of NovoTech Patent Firm, where he helps technology companies build investor‑grade patent portfolios that support fundraising, defensibility, and long‑term competitive advantage. His practice centers on patent strategy, portfolio architecture, and high‑leverage drafting for companies developing AI, machine learning, quantum computing, advanced software‑driven systems, robotics, and other emerging technologies. Babak is also a permanent Adjunct Professor at the University of Maryland, where he teaches Legal Aspects of Entrepreneurship, bringing real‑world IP strategy experience directly into the academic environment. He is a co‑author of the Patent Applications Handbook, published annually by West Publications (Clark Boardman Division) since 1992, and widely used by practitioners as a technical and procedural reference.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
[View Source]