Microsoft & Big Tech want to "own" generative AI: will we & the Feds let them?
The "usual suspects" are determined to squeeze out all AI indies (& control the fate of our content ... and you know what that means)
Welcome to this week’s brAIn dump! As always, first, “the AI:10” - 10 key media-related AI headlines from last week. Next, this week’s “genAI mind-blower” that you gotta watch to believe. Then, “the mAIn event” (my feature headline story). Next, “the cocktAIl” - my special AI event mixology. Finally, “the AI legal case tracker” - updates on the key generative AI copyright infringement cases.
I. the AI:10 - 10 “quick hit” AI headlines from last week
(1) OpenAI and rival Anthropic compete and disagree on just about everything, but on this they agree — no license, consent or compensation is needed to train their AI. Don’t take it from me. Anthropic CEO Dario Amodei said it himself when he spoke the quiet part out loud. Read about it here.
(2) Oops, Adobe did it again! When Adobe released its Firefly AI, it boasted that it was trained only on licensed creative works. That was a major selling point. Little did we know that — actually — Firefly’s AI trained on images from Midjourney, which is embroiled right now in major copyright litigation because it trained its AI on countless unlicensed images (I feature these cases in the AI case tracker below). So, by the transitive property we all learned in middle school, if (a) Midjourney infringed copyrighted works, and (b) Adobe Firefly trained on Midjourney’s images, then (c) Firefly has infringed copyrighted works. It’s elementary! Read about it here.
(3) Its clock may be winding down, but TikTok still moves full steam ahead! Despite Congress’s new bill seeking to ban it, TikTok is developing an advertising tool that generates virtual Influencers who could star in advertising videos. Read about it here.
(4) Speaking of advertising - it’s no joke - brands are adding AI restrictions to agency contracts. And it all makes sense. Why would any brand want to risk getting embroiled in all the ongoing copyright infringement litigation by working with AI companies who trained their AI on unlicensed copyrighted works? Read about it here.
(5) Google wants “in” on the advertising action, of course. It’s latest Workspace app “Vids” gives users the ability to ask AI to make videos for you. Read about it here.
(6) Spotify launches personalized AI playlists and, of course, Jeff Bezos follows suit. Amazon Music launched its own AI playlist generator called “Maestro.” Meanwhile, Euro-based streaming service “Soundtrack Your Brand” announced its own AI music playlist generator for business. The company says that it’s trained on a library of over 100 million commercially licensed songs. Read more here.
(7) You say “responsible AI,” I say “responsible AI” — but which is it? That’s the problem — no two benchmarks for “responsibility” are the same. Read more here.
(8) But it’s not all doom and gloom in AI land (nor should it be). Artists use genAI to create very real, tangible, physical art. Yeah, it’s a thing. Read more here.
(9) EARNINGS ALERT! It’s a big AI week for Big Tech. Meta reports Q1 earnings Wednesday, the 24th and Microsoft and Alphabet follow suit on Thursday, the 25th. Can’t make those earnings calls? Just have your favorite AI chatbot transcribe them.
(10) From the “this has nothing to do with AI, but ….” department. My wife, Luisa, just opened the doors to her new business “The Grounded Mind Collective” - a new hub for young and evolving minds - right on the Pacific Coast Highway in beautiful Encinitas, California (north county San Diego). It’s dedicated to young people in their 20’s and its mission is to give them tools to become purpose-driven leaders, creative entrepreneurs and social impact advocates. The sessions range from grounding, anxiety reduction, and life design — to job-seeking strategies and execution, entrepreneurial/startup coaching, and financial literacy. Check out The Grounded Mind Collective here — and visit the Hub on your next visit to San Diego. Reach out to me to learn more (peter@creativemedia.biz).
II. this week’s genAI mind-blower: just watch this
To start things off, just watch this (and then read below).
This is a demo of Microsoft’s new VASA-1 “lifelike audio-driven talking faces generated in real time” model. All it took was one single image and an audio clip. Microsoft did all the rest. In seconds (of course). As you watch this, and ponder it after, just remember that we are less than 18 months since OpenAI launched ChatGPT. Imagine how things will be 18 months from now. Read more about VASA-1 here (straight from Microsoft itself).
III. the mAIn event - Microsoft and Big Tech want to own the generative AI opportunity: will we and the Feds let them?
The last massive tech-tonic shift that rocked Hollywood – the Internet – birthed whole new categories of companies that ultimately created the Big Tech world we know (and love/fear?) today. Now we’re in the next transformational wave – generative AI – and this time, history most certainly is not repeating itself.
We see no new truly independent players here (at least not yet), and Big Tech is doing everything in its enormous power to keep it that way. And that concentrated power is something about which we in the creative community must be aware. Left unchecked, it both stifles innovation and our ability to place sensible and necessary guardrails. After all, we know how Big Tech feels about our copyrighted works – in its view, there is no reason to pay for it. All’s f-AI-r in AI love and war, after all. Right?
Move beyond the “indie” optics here and think about AI’s realities. OpenAI – a putative indie -- is the leading player. It’s the one that unleashed ChatGPT into a largely unsuspecting world to take the pole position -- and just recently followed with its sensational cinematic quality video generator sequel “Sora” that left much of Hollywood and Madison Avenue in shock and awe. But Big Tech, in the form of its O.G. Microsoft, effectively owns OpenAI.
Sure, OpenAI started its independent life innocently enough (well, as innocent as you can be with Elon Musk as a co-founder). It dedicated itself to serving the public by creating AI to do “good” – in its words, “unconstrained by a need to generate financial return.” But that all changed when the company needed money. Microsoft stepped in with $10 billion, and just like that, principles be damned (Microsoft and OpenAI can’t monetize principles after all)!
Sugar daddy Microsoft’s fingerprints are all over OpenAI. It was Microsoft’s CEO Satya Nadella who came to Sam Altman’s rescue when some on the company’s board (you know, the “do good” ones?) ousted him as OpenAI’s CEO. This was no magnanimous gesture, of course. Nadella had his investment – and first-mover position -- to protect. OpenAI is also dependent upon Microsoft infrastructure to churn all those unlicensed copyrighted works in its AI black box. So much for high ideals. More like, innocence lost.
But there’s even more to this story. Microsoft is also a significant investor in two of OpenAI’s biggest “indie” rivals, Inflection AI and Mistral (over $1 billion in Inception AI alone). So how “indie” are they really? At a very basic level, anytime any company receives a “strategic investment” – as is the case here – that investment comes with strings attached. Strategic investors like Microsoft expect some semblance of control. That’s precisely the point (I’ve led several tech-forward media companies and have experienced this first-hand; it was always a tough decision whether to take investment from “strategics”).
But wait, there’s even more. In a truly magical case of Big Tech sleight of hand, Microsoft recently absorbed Inflection’s entire senior management team into its own, leaving Inflection’s corporate carcass to essentially fend for itself. In other words, it found a way to buy the company without actually buying the company. Why? Because Microsoft knew that the antitrust police likely wouldn’t allow an outright acquisition to happen, especially since the company already finds itself in the FTC’s line of sight. So, Nadella sent his lawyers back to the drawing board, and voila – they came up with a new form of “M&A” without the pesky fuss and muss.
Notice anything else a bit unsettling here? Yes, that’s right – Microsoft is a strategic investor in three of the leading “independent” (cough, cough) generative AI companies. In a classic case of hedging its bets, why just seek control over one when you can control them all (or at least exert some meaningful semblance of control)? Seems like a lot of concentrated power to me.
Now let’s turn to Google, arguably next in line in the generative AI space race. Google, also on the Feds’ “10 most wanted” antitrust list, is playing to win so badly that it’s now openly casting aspersions on OpenAI -- calling out OpenAI’s Sora for stealing YouTube’s IP. As I pointed out last week, the funny thing is that Google is doing the same thing with its YouTube videos – using them to train its AI. Believing that we can’t handle the truth, Google blames its OpenAI derision on OpenAI’s violation of YouTube’s terms of service — rather than using the triggering words of “copyright infringement” (which could come back to bite them, if they did).
Anthropic is another leading Silicon Valley born generative AI independent “unicorn.” But it’s same old, same old here too. Amazon leads the Big Tech investment pack with $4 billion, but Google threw in $2 billion of its own for good measure. So Anthropic certainly is not free from Big Tech’s tentacles. And there we go again – big fish Google, like fellow whale Microsoft, swims in multiple ponds, leaving little room for smaller fish to thrive. They simply don’t have the resources of the magnitude needed to compete effectively right now.
And then there’s Meta, once again a leading player on the Big Tech stage after CEO Mark Zuckerberg jettisoned his bizarro metaverse-ian avatar persona (which is even stiffer than his actual one). Zuck finally found Wall Street religion and refocused the company’s resources on generative AI to catch up fast to the competing Goliaths. When his team determined they couldn’t waste the time or resources necessary to license the content they needed for training, they instead – like his Big Tech brethren – decided, “F’ it! Full steam ahead!” Copyrights be damned. That’s literally just about what his team reportedly said, according to The New York Times.
And let’s not forget formerly invincible Apple, which now faces significant industry headwinds. Just like Meta before it, Apple recently dropped its own earlier “next big thing” dreams -- Tim Cook’s was project Titan, its long-anticipated autonomous car. Now Apple has throttled full speed ahead into generative AI.
So, ladies and gentlemen, that’s your generative AI leader board, starring your usual cast of Big Tech characters that you know and some of you may actually love (especially if they’re in your 401K). But is that a good thing? Is so much power over our AI-impacted creative destinies in the same Big Tech hands what we want?
The Federal Trade Commission is scrutinizing all of this right now. Earlier this year, it launched an official inquiry into Microsoft’s investments into OpenAI and Amazon and Google’s separate investments into Anthropic. But that doesn’t mean the FTC will act or that Big Tech will slow down its investments into genAI “startups” anytime soon. Amazon, in fact, just added $2.75 billion to its original $1.25 billion investment into Anthropic – months after the FTC made its move. And then just last week, the European Union – typically faster and more aggressive in its antitrust pursuits -- cleared Microsoft from formal review of its OpenAI investments.
The more things change, the more things stay the same.
So, what can we do about it? Collective industry advocacy and pressure matter. The issues of copyright and infringement are much higher profile now than they were just a few weeks back.
Ultimately, the fundamental issue about whether AI training on copyrighted works without consent and compensation constitutes infringement or is, instead, defensible as a “fair use” will wind its way to the U.S. Supreme Court. And even this wacky Court will rule in favor of the creative community and find infringement, based on its recent precedent in the Andy Warhol/Prince case. Read my full analysis about this here.
What do you think? Send me your feedback and thoughts at peter@creativemedia.biz.
IV. the cocktAIl- your AI mix of “must attend” AI events
After all, it’s always happy hour somewhere!
(1) My next Digital Hollywood virtual AI roundtable is THIS THURSDAY April 25th, 12 noon Pacific (3 pm Eastern). It’s all about generative AI and Video this time — and how OpenAI’s “Sora” and other AI video generators will transform both Hollywood and Madison Avenue advertising. Last time I interviewed The Police’s Stewart Copeland about AI and music. This month’s roundtable features a great “show and tell” — both a live demo of one of the leading generative AI video tools and leading media/marketing/tech execs who have used it. The roundtable is absolutely free. And absolutely fascinating … and important. Register now.
(2) AI LA’s big “A.I. on the Lot” event is less than one month away on May 16th! It’s a “must attend” event (I’ll be there). Get 20% off now when you register using promo code “PETER” as you check out. Register via this link. Let me know if you want to meet there. Reach out to me at peter@creativemedia.biz.
(3) Check out Digital Hollywood’s first generative AI-focused virtual summit, “The Digital Hollywood AI Summer Summit,” coming soon on July 22nd - 25th. The sessions are comprehensive and outstanding. I’ll be moderating two of them. Learn more here via this link.
check out Creative Media and our AI-focused services
V. the AI legal case tracker - updates on key AI litigation
Rather than lay out the facts of each case - and the latest developments - in every newsletter, click on this “AI case tracker” tab on “the brAIn” website. You’ll get all the up-to-date information you need (including my detailed analysis of each). These are the cases I track (several important developments this past week).
(1) The New York Times v. Microsoft & OpenAI
(2) Sarah Silverman, et al. v. Meta
(3) Sarah Silverman v. OpenAI
(4) Universal Music Group, et al. v. Anthropic
(5) Getty Images v. Stability AI and Midjourney