DeepSeek: What It Means for Media & Entertainment
It's Not Just a Silicon Valley Stunner. Hollywood's Potential Ramifications Are Massive Too.
Welcome to my latest “brAIn” dump. Suffice it to say, I have AI-lot on my mind! We start off with “But First” — your “must know” AI/media news story of the week (major copyright news). Then, it’s the“mAIn event” — my feature about what DeepSeek’s shot across Silicon Valley’s AI bow means for media and entertainment (hint: a lot!). Next, my AI-xpert of the week — Andy Beach, who just left Microsoft (CTO, media/entertainment) gives his 90-second “take” of DeepSeek (big deal or not?). And, the “mosAIc” — a collage of key AI stories that impact media, and my picks of “must listen” podcast episodes about DeepSeek & more). Finally, the “AI Litigation Tracker” — updates on key GenAI/media cases by law firm McKool Smith (you can also check out the “Tracker” via this link here).
I. But First …
The Copyright Office Finally Issued Its Highly Anticipated Guidance About AI & Copyrightability
At long last — and after receiving 10,000+ submissions to its request for comment (about which I wrote a couple newsletters back) — the U.S. Copyright Office just-issued its highly anticipated guidance on the copyrightability of generative AI infused creative works. The Office’s report addresses this fundamental issue specifically in the context of Hollywood production, post-production and overall workflows. Where does copyright protection for AI-enhanced works begin and end? And what are the real world implications for the Arts, the individuals who power it, and the studios that bankroll it? It’s a mixed bag that attempts to please everyone, but adds little real world clarity. Entertainment trade pub Variety wrote an excellent “must read” summary of it all (here’s the link). And AI/media specialist Avery Williams, a Partner at leading law firm McKool Smith, gives his expert legal “take” in “the mosAIc” below. IMPORTANT NOTE: the Copyright Office’s report does not discuss the fundamental issue of infringement v. fair use when LLMs train on copyrighted works. That specific guidance is expected later this year.
II. The mAIn event - DeepSeek Deep Sixes U.S. GenAI Players. So What Does It Mean for Media & Entertainment?
Listen to my latest podcast episode that discusses my article (generated by Google NotebookLM - which faithfully discusses my points and gives helpful perspective).
Intro
Over the years, China hasn’t exactly been a bastion of respect for U.S. copyrighted works. Now, even U.S. Big Tech feels China’s disrespect. By now, all of you — not to mention your 401 K plans — know about DeepSeek, China’s new generative AI power tool that’s been #1 in the App Store for days (2+ million downloads in just 7 days) and called into question Silicon Valley GenAI’s fundamental economics. DeepSeek’s headline story is that it just released an AI chatbot that rivals OpenAI … but at 1/1000th of the cost. Obviously, if true, this has massive ramifications that are beginning to play out in real time. (See Section III below for an expert’s take).
What DeepSeek Means for Media & Entertainment
Here’s the headline story for media and entertainment. If the creative community thought it was difficult before DeepSeek to establish basic guardrails on GenAI’s training on its copyrighted works (not to mention deep fakes and other privacy & NILV issues), just imagine the likelihood of any success now? DeepSeek has essentially “deep sixed” the likelihood of any substantive legislation or regulation coming out of Capitol Hill. To be fair, I was already skeptical that any real federal legislation in support of Hollywood was forthcoming. Just last week I wrote about Mr. T’s photo opp with Sam Altman and other AI titans about their new $500 billion AI infrastructure bill — and the “damn the torpedos” ethos it represented.
Well, what a difference a week makes! Now, with DeepSeek, maybe that $500 billion should instead be $500 million (i.e., 1/1000th)? A downright bargain, yes. But maybe not music to the ears of Nvidia and GenAI’s major players. But I digress …
Make no mistake. DeepSeek is a “call to arms” (literally an AI arms race) by Donald’s tech whisperers — Elon, Zuck & their brethren — all of whom undoubtedly are telling their new boss that U.S. AI must be unleashed at all costs. In the name of U.S. tech dominance and U.S. national security.
This newly highly charged AI Accelerationism puts the media and entertainment industry at Def-Con 1 on the GenAI risk-ometer. Remember, Creators and Rights-holders expect, both legally and ethically, fair consent & compensation, which seems pretty fair to me. But Creators, Congress will not help you on this righteous path.
So What Can Media & U.S. Tech Do About It?
The industry must help itself and explain that its demands are not about slowing down GenAI development. It’s simply about stopping the unconsented and uncompensated “taking” of its creative works. After all, OpenAi says that its tech can’t simply be “taken” by DeepSeek (see Section IV below for more about this intriguing and very ironic development).
Tech’s great “taking” has led to the ongoing 40+ copyright and related media infringement litigations now winding their way in the courts. That litigation means friction. And friction holds back movement. So if the goal here is to unleash U.S. AI development to compete (and win) versus new threats like DeepSeek, then it’s in Big Tech’s interests to eradicate that friction. And the best way to do that is through cooperation with the creative community. That means that licensing solutions between GenAI developers and media companies and creators are now more critical than ever. That means there’s no time to waste.
Those Billions Saved Should Be Used for Licensing
If DeepSeek’s basic shot across the bow is real — i.e., that a fraction of what was previously believed to be necessary for GenAI development is needed — then those hundreds of billions of dollars of savings (for funds otherwise allocated for AI infrastructure, like Meta’s $65 billion for this year) can and should be used, at least in meaningful part, to fast-track content licensing discussions. Accelerated licensing of trusted content sources — i.e., ethical AI - could also significantly differentiate U.S. AI developers from Chinese counterparts, many of which historically haven’t been known to respect U.S. intellectual property.
Big Tech — let’s form allies to win and achieve your goals faster, and better. The media and entertainment industry can be your allies.
As Hollywood agent Jerry Maguire would say, “Help us, help you.”
(For those of you interested in learning more or exploring licensing opportunities, reach out to me at peter@creativemedia.biz).
III. AI-expert Andy Beach, former CTO Microsoft: DeepSeek & What It Represents (in 90-Seconds!)
Andy Beach, until just weeks ago, was CTO of Microsoft’s Media & Entertainment group. He’s a leading voice in the world of GenAI — helping to explain AI’s promise, as well as its limitations and perils to Hollywood. In this short video below, Andy gives his candid, independent thoughts about DeepSeek (and how groundbreaking he feels it really is). (Just click on this thumbnail below to watch).
NOTE: I’ll feature my full exclusive conversation with Andy about generative AI and Hollywood in next week’s newsletter.
IV. The mosAIc — More AI News That Matters
(1) Judge Says: “OpenAI, Shine The Light!”
Everything you always wanted to know about OpenAI’s training data set, but were afraid to ask! Actually, the media and entertainment world asked. A lot. And the answer? Some rendition of “publicly available data,” whatever that means (but … wink wink, nudge nudge … we all kinda’ know what it means). And now, presumably, we’ll know exactly what it means if federal Magistrate Judge Robert M. Illman gets his way. He just ordered OpenAI to reveal what “data” it used to train its flagship GPT-4 model. Read more about it here via Bloomberg.
(2) OpenAI Says: “DeepSeek, You Can’t Just Take!”
This past week, OpenAI accused DeepSeek of “stealing” its core AI tech to develop its new rivaling, and threatening, LLM. An OpenAI spokesperson was quoted as saying, "We take aggressive, proactive countermeasures to protect our technology and will continue working closely with the U.S. government to protect the most capable models being built here." Hey fellow citizens of the creative community — inspired by Alannis Morissette’s classic song — “isn’t it ironic?” Read more about it via Axios.
(3) So What Does the Copyright Office’s New GenAI Guidance Mean? Lawyer Avery Williams Breaks It Down
I asked Avery Williams, Partner at the law firm of McKool Smith who specializes in AI and copyright matters, to give his thoughts about the Copyright Office’s new GenAI guidance. This is what he tells me.
The Copyright Office’s position that prompting alone is currently insufficient to convey authorship of the AI-generated work would render hundreds of millions of images essentially authorless. Courts have split on the question of whether, through iterative prompt engineering and image selection, a prompter could establish enough creative control to be considered an author. If courts adopt The Copyright Office’s position, it will clarify what does not qualify for copyright protection, but would leave many AI-generated works unprotected, even if substantial effort had gone into iterative prompting and generation. From there, the Copyright Office’s analysis is unsurprising. Elements of an AI image that reflect a copyrighted input image are protectable. Transformative or creative arrangement of AI-generated elements may be protectable. And modifications to AI works may be protectable if the creative input is sufficient. Predictability is generally good for the law, but it remains to be seen if courts are willing to consider prompting as a form of creative input.
Thanks Avery! If you’d like me to connect you to Avery to discuss this or other matters, reach out to me at peter@creativemedia.biz.
(4) Podcast Episode of the Week (About DeepSeek)
I urge you to listen to a great overview and discussion of DeepSeek by one of my “go to” podcasts, Hard Fork (via The New York Times). How can you resist an episode titled, “Your Guide to the DeepSeek Freakout” (you can find it here).
(5) Book of the Week (Reid Hoffman’s Optimistic “Take” on AI & What to Expect)
I’ve only just begun reading (listening, actually) to the LinkedIn founder’s new book titled “Superagency: What Could Possibly Go Right with Our AI Future” (check it out here). It’s always important to hear Big Tech’s frequently very different perspective about AI, and Hoffman is a leading voice of course. I certainly don’t always agree with him. But his new book is smart. Fascinating.
Follow me on BlueSky via this link.
You can also continue to follow my longer daily posts on LinkedIn via this link.
V. AI Litigation Tracker: Updates on Key Generative AI/Media Cases (by McKool Smith)
Partner Avery Williams and the team at McKool Smith (named “Plaintiff IP Firm of the Year” by The National Law Journal) lay out the facts of — and latest critical developments in — the key generative AI/media litigation cases listed below. All those detailed updates can be accessed via this link to the “AI Litigation Tracker”.
(1) Raw Story Media v. OpenAI
(2) The Center for Investigative Reporting v. OpenAI
(3) Dow Jones, et al. v. Perplexity AI
(4) The New York Times v. Microsoft & OpenAI
(5) Sarah Silverman v. OpenAI (class action)
(6) Sarah Silverman, et al. v. Meta (class action)
(7) UMG Recordings v. Suno
(8) UMG Recordings v. Uncharted Labs (d/b/a Udio)
(9) Getty Images v. Stability AI and Midjourney
(10) Universal Music Group, et al. v. Anthropic
(11) Sarah Anderson v. Stability AI
(12) Authors Guild et al. v. OpenAI
NOTE: Go to the “AI Litigation Tracker” tab at the top of “the brAIn” website for the full discussions and analyses of these and other key generative AI/media litigations. And reach out to me, Peter Csathy (peter@creativemedia.biz), if you would like to be connected to McKool Smith) to discuss these and other legal and litigation issues. I’ll make the introduction.
About My Firm Creative Media
My firm and I specialize in market-defining strategy and content licensing for generative AI, breakthrough business development and M&A, and cost-effective legal services in the worlds of media, entertainment, AI and tech. We develop game-changing strategic opportunities and leverage our uniquely deep relationships to reach key decision-makers and influencers in record time to execute. Not just talk.
Among other things, we represent media companies for generative AI content licensing, with deep relationships and market insights and intelligence second to none. Reach out to Peter Csathy at peter@creativemedia.biz if you’d like to explore working with us.
Send your feedback to me and my newsletter via peter@creativemedia.biz.