Lauri Donahue, Director of Legal Content at LawGeex
As we blogged about back in the day,
Many people make others sign non-disclosure agreements (NDAs) before they reveal their business ideas to them. They assume that an NDA will prevent the other party from using or revealing their ideas.
But NDAs aren’t like the magical protective spells some people seem to think they are.
In fact, an NDA could be worthless if the party disclosing information isn’t willing to go to court to enforce it:
People can — and do — violate NDAs. They aren’t scared off by the threat of litigation, because they know it’s often an empty threat.
The average startup, small business, or freelancer may not be willing or able to invest in a lawsuit to enforce an NDA. Unless it’s a “bet the company” case, they may just decide to suck it up when someone steals their ideas.
Another reason that NDAs may be worthless in protecting “ideas” is that many “ideas” don’t qualify for protection under an NDA.
This is illustrated by a recent case involving blockchain and CryptoKitties.
A blockchain, for the uninitiated, is an online distributed ledger system. It’s like a spreadsheet that exists on many different computers at the same time. When one copy is updated, all copies are updated.
The best-known application of blockchain is virtual currency (like Bitcoin), but there are many others — including virtual cats.
CryptoKitties are the brainkittens of Launch Labs, a Canadian company doing business as Axiom Zen, which raised $12 million in venture capital investment.
The CryptoKitties game uses Ethereum blockchain technology to allow users to “securely buy, sell, trade, and breed genetically unique virtual cats.”
As Motherboard explains,
The genome of each cat consists of a 256-bit number and breeding basically involves mashing these numbers together to produce a new 256-bit number that represents a new cat with unique physical characteristics.
After a smart contract is registered on the blockchain, the designated cat’s status is changed to “pregnant.”
Giving birth to a virtual kitten requires another smart contract — and real money. Last December, the “birthing fee” quickly increased from the equivalent of 90 cents to the equivalent of $6.69.
As of December of 2017, the median price of a virtual kitten was $25.04, according to CNBC.
As CNBC explains,
The primary distinguishing feature [between kitties] is “cooldown” time, or the time it takes a kitty to recover after breeding. These vary from “fast” at 1 minute to “Catatonic” at 1 week and “will increase each time the Kitty breeds.”
As with Beanie Babies, the market “value” of CryptoKitties derives from their relative scarcity and a FOMO feeding frenzy.
The game launched in November of 2017. Within a month, 180,000 people were playing, and the game was bringing in $20 million in revenue per month.
“Only” four billion unique cats can be bred. In December of 2017, a virtual cat called Genesis sold for about US$117,000.
The NDA battle over “Celebrity Kitties”
Even before the game launched, Axiom Zen’s product manager told Vice.com in October of 2017 of plans for celebrity kitties:
We’re the flag bearer for the rest of this ecosystem in terms of getting celebrities to associate with kitties and the Ethereum network.
And now we get to the NDA part…
In February of 2018, Jevon Feinblatt, the CEO of a company called Founder Starcoin, emailed an Axiom Zen employee looking for a business partnership.
Founder Starcoin is a San Diego company that operates in the “cryptocollectibles” market.
Jody Rebak, Axiom Zen’s chief of staff, asked Feinblatt for more information about what he had in mind.
Feinblatt then proposed that the companies sign a mutual non-disclosure agreement (mutual NDA).
The NDA defined “Confidential Information” as “any nonpublic information, technical data or know-how.'”
After the NDA was signed, Feinblatt sent Axiom Zen a slide deck describing a proposal for how celebrities could
commoditize themselves, fund their projects, and create a new type of asset via the first security complaint [sic] token platform for entertainers.
The deck included an image of NBA star Stephen Curry.
The parties exchanged a few emails but had no substantive discussions of the idea.
In May, Axiom Zen released three “CurryKitties” based on Stephen Curry.
Feinblatt promptly sued, claiming that Axiom Zen had misappropriated his trade secrets, and sought a preliminary injunction.
In his moving papers, he characterized his “secret” as “licensing digital collectibles based on athletes, entertainers, and celebrities.”
At the hearing on the injunction, Axiom Zen argued that Feinblatt’s alleged “secret” was already in the public domain. The alleged “secret,” claimed Axiom Zen, was nothing more than the idea of using celebrities to sell products — which is a very old idea indeed.
In addition, Axiom Zen had publicly announced plans to develop celebrity-themed kitty products months before Feinblatt sent his deck.
Very commonly, NDAs exclude certain types of information from the definition of “confidential information.”
Confidential Information shall include all information provided hereunder in writing, orally, visually and/or observed while on the premises of either party, except any portion thereof which:
- is known to the recipient, as evidenced by its written records, prior to receipt thereof under this Agreement;
- is disclosed to the recipient by a third person after the full execution of this Agreement, if that third person has a legal right to make such disclosure;
- is or becomes part of the public domain other than through breach of this Agreement by recipient; or
- is independently developed by or for the recipient as evidenced by its written records, without reference to Confidential Information received from the disclosing party.
However, information in the public domain is already excluded from the definition of a trade secret as a matter of law.
As the US Supreme Court said in Ruckelshaus v. Monsanto Co., and as the court quoted in the CryptoKitty case, “Information that is public knowledge or that is generally known in an industry cannot be a trade secret.”
Although that case applies to trade secrets specifically and not confidential information generally, the same logic would seem to apply to all “confidential” information: how can something be “confidential” (and thus protected by an NDA) if it’s already widely-known?
But somewhere along the line, some lawyer decided to add the “public domain” exception to the definition of “confidential information” in an NDA. And now it’s a ubiquitous feature of NDAs.
As the court noted, the fact that the CurryKitty was allegedly “the world’s first officially licensed sports cryptocollectible” didn’t make the idea a trade secret:
Marrying the concept of celebrity licensing with blockchain technology appears, on its face, to be unremarkable, obvious, and general knowledge. Nearly every industry attempts to gain celebrity endorsements for products…
Celebrity licensing is generally known and Plaintiff has not demonstrated how marrying this idea with the blockchain industry is a secret.
As noted above, many NDAs exclude from the definition of “confidential information” information that the receiving party independently develops.
Again, this exception already exists in the law when it comes to trade secrets. The Uniform Trade Secrets Act (which is the model for most state laws on trade secrets)
does not prevent a person from using independently developed or properly obtained trade secret information already in the possession of another.
Additionally, the federal Defend Trade Secrets Act (DTSA) defines trade secrets as
information deriv[ing] independent economic value . . . from not being generally known to, and not being readily ascertainable through proper means by, another person who can obtain economic value from the disclosure or use of the information.[Emphasis added.]
So again, it seems questionable whether NDAs really need this exclusion.
In the CryptoKitty case, the court found that Axiom Zen was starting to negotiate to use Curry’s likeness a month before Feinblatt sent his deck with Curry’s picture.
As the court noted,
The evidence demonstrates that Defendant, not Plaintiff, developed the idea to license digital collectibles using the likeness of celebrities first…
Because Defendant arrived at its business model independent from and before receiving Plaintiff’s slide deck, Defendant could not have appropriated Plaintiff’s trade secrets.
If you want to read more, the case is Founder Starcoin v. Launch Labs, Inc.
Everyone who’s ever taken a shower has had an idea
Ideas are over-valued — especially by their creators.
Nolan Bushnell, the co-founder of Atari, once said,
Everyone who’s ever taken a shower has an idea. It’s the person who gets out of the shower, dries off and does something about it who makes a difference.
And as the New York Times noted, quoting angel investor Chris Schultz,
Everyone thinks their idea is extremely unique, but the idea is really 1 percent of the value. The value is in the execution.
Many venture capitalists and other investors are reluctant to sign — or outright refuse to sign — NDAs because they deal with “ideas” all the time and don’t want to get entangled in a trade secret “theft” catfight… like the CryptoKitty case.
But that doesn’t mean that investors are out there stealing “ideas” right and left.
Paul A. Jones, co-chairman of the venture best practice group at the law firm Michael Best & Friedrich, told the Times, “I’ve never known a reputable investor to steal an idea and create a company around it.”
The risk of “idea theft” has to be balanced against other risks — like the risk of losing the first-mover advantage.
Victor W. Hwang, chief executive of T2 Venture Creation, an investment firm in Portola Valley, California, told the Times,
In the life of a start-up company, you might have to sign 30 to 50 N.D.A.s. That’s a week each time and a year of holdups. The risk of going slow is bigger than the risk of being copied.
Speeding up NDA Review
Corporate law departments and outside law firms do often take a week (or even more) to review standard NDAs.
The actual review may take less than an hour because NDAs are usually short, but it can take a week or more to get around to it.
It doesn’t have to take so long.
In a landmark study LawGeex conducted with law professors from Duke and Stanford, the LawGeex AI reviewed five NDAs in 26 seconds.
Also, the LawGeex AI achieved an accuracy level of 94% — compared to 85% on average for 20 experienced lawyers.
Again, an NDA isn’t a magic wand for protecting ideas.
But if signing an NDA is required to move a deal forward, it’s a good idea to get it reviewed as quickly as possible.
The LawGeex AI isn’t magic, either — but as Arthur C. Clarke said, “Any sufficiently advanced technology is indistinguishable from magic.”
For a demonstration, please contact us.