附近上门

Artificial intelligence Media & entertainment tech

Can We Please Still Have Nice Things That AI Hasn鈥檛 Touched?

Illustration of robot drawing in the air. (Generative AI)

鈥淭he Wizard of Oz鈥 鈥 yes, that Wizard of Oz 鈥 is . Cloud and , in collaboration with the Sphere in Las Vegas, have used something they鈥檙e calling 鈥減erformance generation鈥 and 鈥渙utpainting鈥 to enhance the film鈥檚 resolution and extend the background beyond what was ever shot.

According to Ravi Rajamani, managing director of generative AI engineering at , AI has 鈥渢ouched over 90% of the movie.鈥

Ninety. Percent.

Everyone is fawning over it, which, yeah, I get. We can all agree that technology is amazing. We鈥檝e all marveled at what AI can do: generate realistic images, compose music, write essays, predict weather patterns, help doctors detect diseases, and so on.

But when , CEO of Google Cloud, says this isn鈥檛 really a cinematic experience, but an 鈥渆xperiential鈥 one, I can鈥檛 help but sigh and think: Do we really need AI to touch everything?

Because some things 鈥 some beautiful, classic and already timeless things 鈥 are better left untouched. 鈥淭he Wizard of Oz鈥 is not just a movie. It is, for tens of millions of people, a memory, a mood and a moment. It鈥檚 that sepia-toned Kansas that bursts into technicolor magic. It鈥檚 Judy Garland鈥檚 voice trembling with hope as she sings 鈥淥ver the Rainbow.鈥 It鈥檚 the Tin Man rusting mid-sentence and the Cowardly Lion bawling like a baby. It鈥檚 the sound of ruby slippers clicking down a yellow brick road, and it is 鈥 crucially 鈥 enough as it is.

And now AI wants to “outpaint” it? Extend the background? Fill in scenery that never existed? It’s like someone taking a classic oil painting and saying, 鈥淵ou know what this Monet really needs? A few extra lily pads generated by machine learning.鈥

Pay no attention to the AI behind the curtain

I鈥檓 thrilled when AI helps people solve complex problems, enables creativity in new ways, or makes life easier for those who need it.

But there鈥檚 a difference between innovation and intrusion. There鈥檚 a line between enhancement and overreach. Somewhere along the way, tech giants like Google seem to have forgotten that not everything needs to be optimized, upscaled or AI-enhanced.

Some things are sacred. 鈥淭he Wizard of Oz鈥 is one of them.

I understand the argument: this is a new kind of experience. The Sphere isn鈥檛 just a theater, it鈥檚 a dome of digital immersion. You鈥檙e not just watching a movie, you鈥檙e 鈥渆xperiencing鈥 it with full surround visuals, sensory stimulation and a production pipeline that would probably confuse even the Scarecrow with a brain.

But does that mean we have to rewrite and reskin every classic to fit into this new paradigm? What鈥檚 wrong with screening the original film as-is?

Let it breathe. Let it play, untouched, for new generations to experience the same magic that鈥檚 captivated audiences for nearly a century. Imagine watching Dorothy step into Oz in that original 1939 technicolor glory, in a dome that simply celebrates the original rather than trying to reinvent it.

Instead, it feels like the industry is treating classic art as raw material for whatever the latest tech buzzword is. This time it鈥檚 鈥減erformance generation.鈥 Next time, it鈥檒l be something else.

We鈥檙e already seeing AI-generated actors in ads, AI-written scripts and deepfakes in entertainment. Are we heading toward a future where no piece of media is ever considered finished? Where even the most iconic stories are endlessly retooled, reprocessed and reimagined by machines until the original is a ghost of itself?

There鈥檚 something particularly ironic about 鈥淭he Wizard of Oz鈥 being the test subject here. A story that鈥檚 about the power of imagination, about finding your way home, about the heart, brains and courage inside us 鈥 not inside a data center in Silicon Valley. It鈥檚 a deeply human film. It鈥檚 messy and charming and grounded in emotion and theatricality and, yes, the limits of 1939 production technology.

But those limits are part of the charm. The painted backdrops, the practical effects and the costumes don鈥檛 detract from the story. They enhance it, because they remind us that real people made this.

I鈥檓 not saying we should freeze art in time and never touch it. There鈥檚 value in restoration, remastering and making things accessible to new formats and audiences. But there鈥檚 a difference between polishing a gem and melting it down to make a whole new piece of jewelry.

This version of Oz feels like the latter.

Where does the yellow brick road of AI lead us?

I also worry about the precedent this sets. Today it鈥檚 鈥淭he Wizard of Oz,鈥 tomorrow 鈥淐asablanca鈥 with AI-generated colorization and new dialogue, next week, 鈥淭he Godfather鈥 with new story arcs and actors reverse-aged by AI. Eventually, are we going to let algorithms rewrite 鈥淐itizen Kane鈥 because some machine learning model predicts a different ending would have more 鈥渆motional engagement per second?鈥

Let me be clear: it鈥檚 not that this AI remake of Oz won鈥檛 be cool. It probably will be. I鈥檓 sure it鈥檒l be wild to see flying monkeys in 360 degrees. But cool doesn鈥檛 always mean necessary, and it definitely doesn鈥檛 mean better.

There鈥檚 a certain arrogance in thinking everything needs to be remade through today鈥檚 tech, as if the original isn鈥檛 enough unless it鈥檚 been algorithmically enhanced. But some stories are powerful because of their limitations, their imagination and the memories they hold. We don鈥檛 need AI to re-thread what already connects generations.

So go ahead, Google 鈥 build your wonderland. But remember: Sometimes the magic is already there 鈥斅爊o need to outpaint it.


is the chief strategy officer for . He holds a law degree and has taught entrepreneurship at and the , and was elected to Fastcase 50, recognizing the top 50 legal innovators in the world. His writing has been featured in , , , , , , , and many other publications. He was nominated for a Pulitzer Prize .

Related reading:

Illustration:

Stay up to date with recent funding rounds, acquisitions, and more with the 附近上门 Daily.

67.1K Followers

CTA

Discover and act on private market opportunities with predictive company intelligence.

Copy link