Table of Contents[Hide][Show]
Take-Two CEO Strauss Zelnick is surprisingly upbeat about generative AI in the video game industry.
In an event at the Paley Center for Media, Zelnick shared what he thought generative AI is:
You have to look at it as a tool set. Our business has been involved with digital tools since its inception.
I think we’ll probably be able to create a bunch of efficiency, and we’re already trying to do that.
We’re reviewing 200 different opportunities at any given time, but what AI is not going to do—because remember, what AI is, is the combination of big data sets with a bunch of computers within a natural LLM, a large language model.
And by definition, a data set is backward-looking, while creativity is forward-looking. To the extent that AI appears to be forward-looking, it is what a predictive model is.
OK, But What Is He Talking About?
We want to take a step back and disentangle what Strauss said here. The common perception is that modern AI is trying to replace jobs, especially creative work, by copying human work and doing a poor job of it.
When Strauss refers to AI as a digital tool, he refers to tools that have existed before ChatGPT and OpenAI. We all know AI has been around since the start of video games.
Even the most primitive text based adventure games like Zork have a primitive form of AI running in them. It had to so that you could type instructions in your own words and it would be able to interpret them.
Strauss also clearly makes a distinction between that old AI and the newer kind that’s fueling the AI boom. He mentions large language models and predictive models.
Most interestingly, he criticizes this new AI as ‘backward-looking.’ It’s a criticism that’s borne from actual experience testing the technology, so he can’t be accused of just being technophobic.
Where Strauss Thinks AI Will Be Useful
Strauss also talked about what he thinks AI will be able to do for the game industry:
It will not reduce employment, it will increase employment. Technology always increases productivity, which in turn increases GDP, which in turn increases employment.
And you know, in 1865, 65% of the US workforce was involved in agriculture. Today, we produce food for America and the rest of the world, and 2% of the workforce is involved in agriculture.
And I defy you to find anyone who recently said to you, ‘It’s so horrible I can’t get a job as a farmer,’ and employment is way up.
AI is a great thing. It’s a great thing for every industry.
Will it create a genius? No. Will it create hits? No. It’s a bunch of data with a bunch of computers with a language model attached.
Strauss seems realistic about what AI is actually able to do, and he sees through the most inflated hype like the rest of us do.
But it seems he also now sees what applications will come out of the AI boom. It seems strange to be so optimistic about AI with the cloud of negativity surrounding it.
But Strauss seems to be someone who already sees the endgame, and is already moving his company to get there. We can only hope his predictions about AI and gaming’s future prove true.
