More intelligence of the artificial variety

Early on in my brilliant career, a number of my colleagues (including my manager, someone who reported to me, and my best work bud Peter who was and still is one of my closest friends) joined a fancy AI startup in Cambridge. The product was a system  - which ran on a pricey LISP machine -  that helped businesses make capital budgeting systems. Basically, it was MIT/Sloan Professor Stewart Myers textbook in a box. It was going to replace the existing method of capital budgeting, which was basically using a spreadsheet to grub numbers (I believe Lotus 1-2-3 was the top dog at that time, Microsoft's Excel just being introduced) and then overriding the numbers with gut instinct, with gut instinct often worked out on the back of a cocktail napkin. And it was going to result in better decisions than the existing methods of capital budgeting. And it was going to cost a shit-ton of money. 

I interviewed there, but was weighed and found wanting and didn't get an offer. The company was known for hiring the best and the brightest, so I was a bit miffed - okay, I was a shit-ton miffed - that they turned me down. My wonderful friend Peter assured me that it was because during the interview process, I asked one too many pointed questions. I know he was just trying to make me less shit-ton miffed, but I had asked the head guy whether they thought there were many companies that were going to spend $150K to replace something they were doing on a cocktail napkin for free. So there was that.

In letting me down, I was told that they weren't going to make me an offer because they didn't feel that I was "emotionally ready to leave" the company I was working for.

Anyway, it does seem that I'm enough of a miff-holder to have been just delighted when this outfit went tits up. (I'll show you "emotionally ready.")

I haven't thought about this company in years, but it ear-wormed its way into may mind when I saw an article on a company, Cinelytics, that has an AI-driven system to help movie studios decided whether to green light a project.

And they just landed Warner Bros. as a client.

Under the new deal, Warners will leverage the system’s comprehensive data and predictive analytics to guide decision-making at the greenlight stage. The integrated online platform can assess the value of a star in any territory and how much a film is expected to make in theaters and on other ancillary streams.

...While the platform won’t necessarily predict what will be the next $1 billion surprise, like Warners’ hit Joker, it will reduce the amount of time executives spend on low-value, repetitive tasks and instead give them better dollar-figure parameters for packaging, marketing and distribution decisions, including release dates. (Source: Hollywood Reporter)

Ah, if only I had a dollar for every time I made the marketing argument that whatever software I was plumping for was going to free folks up to focus on higher-value tasks...

Anyway, AI has come a long way since the mid-1980's when the best it could offer was a textbook in a box. Nonetheless, Hollywood has been reluctant to adopt AI, as it "fancies itself as a town that operates on gut instinct rather than algorithms, for better or for worse."

Cinelytics found Tobias Queisser plays down the notion that AI will replace all that gut instinct magic - the sort of gut instinct that greenlights making a movie out of, say, Cats. (Which was Universal's fiasco, not Warner Bros.')

“Artificial intelligence sounds scary. But right now, an AI cannot make any creative decisions,” says Queisser. “What it is good at is crunching numbers and breaking down huge data sets and showing patterns that would not be visible to humans. But for creative decision-making, you still need experience and gut instinct.”

That gut instinct has held Warner Bros. in pretty good stead over the years. The Jazz Singer. Public Enemy. Now Voyager. The Maltese Falcon. Cartoon "stars" like Daffy Duck and Bugs Bunny. TV fare like Maverick and Buffy the Vampire Slayer. Blockbusters like The Dark Knight and the Harry Potter films. 

On the other hand, maybe Cinelytics could have helped them avoid a megaflop like King Arthur: Legend of the Sword, which lost a whopping $150M. 

Hope AI never fully replaces creative decisions, but I wouldn't bet against it.

Meanwhile, I noticed that there's a Harvard Business School case on that software company that so miffed me. I think I know how I can get my hands on a copy of it. Might be a hoot to read all about it.

•