AI & Data Licensing
Last week, writing about copyright and IP under assault by big tech, our industry under existential threat, Thomson Reuters vs. Ross potentially — too little too late, the big moves just keep coming.
We are hearing from and advising clients who are finding their IP all over heck and back and who are trying to put the genie back in the bottle.
Memo to industry: We are never going back.
Our data is everywhere, good, bad, and otherwise. It’s on the open web, the dark web, and in LLMs that it shouldn’t be. For now, the only way to really deal with this is to enforce rights. I’m not a lawyer, but as a businesswoman, it’s clear to me, just like with trademarks and service marks — if one doesn’t stand up for them each and every time, it gets harder to do so down the line. One has to be a he-lion or she-lion, as the case may be. And with AI, this is incredibly important. Don’t leave the door open for a sliver of ‘ok-ness’ for unlicensed use.
We see companies that:
- Don’t have a licensing strategy for AI. (Full disclosure we work with clients to create those licensing strategies and are good at it. (Call us if you need assistance.)
- Miss centralizing the mechanisms and methods by which they do license so there is a coordinated process, a go-to inside various functions, and coordinated activity in licensing.
- Don’t treat data, content, analysis, editorial — whatever you wish to call it — as money in the bank. Companies don’t let anyone sign checks, distribute cash, or authorize wires. There are authority levels and signatories. We need to treat data, content, analysis, editorial — like the asset it is, especially our money in the bank. It is that important in companies like ours.
- Are not clear on their own rights nor details about the content they have. Imagine not knowing what cash is in the bank, what the contours of A/R, A/P are — what we have rights to and can collect and what we don’t.
My colleague, David Worlock, has been looking at newcos focused on licensing and rights and they are popping up. This is our moment to shout out CCC and CLA and others like them who do the yeoman’s and yeowoman’s work to enable legitimate access to quality content.
Like good water in the pipes — quality matters, and good data in AI is no different. David cited some interesting firms we are evaluating who operate in this developing sphere and the topic continues to go mainstream because good data matters in the next generation of tech.
Big tech will have you think it’s all on the open web (peak data + synthetic data) or that they have an inherent right to gobble it up. IP owners know better and must enforce this each and every day. The world is changing at Mach-10 speed. Look no further than this week’s headlines about Regeneron acquiring 23andMe assets, including its liquid gold — its database. So much for the rights anyone signed away.
While Mr. Musk wants us to use brain implants, and Mr. Zuckerberg wants us to wear AI glasses, Mr. Altman and Mr. Ivy have new big ideas, and I can guarantee you it’s about AI everywhere all the time. Our kids can’t turn off their devices and have anxiety from being online 24/7, but big tech wants us always on and is now building the next generation devices to make it so. Good data better run in those pipes; it’s up to us to make it so.