Following last fall’s announcement of Electronic Arts’ big take-private deal, the video game publisher made another splashy move by revealing a multi-year partnership with James Cameron-backed Stability AI.
“The Sims,” ”Battlefield 6″ and “Madden NFL” maker EA also made a strategic investment in stability AI in its latest funding round in October, and although it’s been several months since the team-up, both companies still have a long way to go before gamers see the outcome of the deal.
Everything is still in the very early stages, says Mihir Vaidya, chief strategy officer at EA. variety In a January 16 interview, he said that conversations about a partnership between EA and Stability AI began long before EA was courted by Saudi Arabia’s Public Investment Fund (PIF), Silver Lake, and Jared Kushner’s Affinity Partners for a $55 billion take private.
“I think one of the things that we’re very proud of at EA is that throughout our history we’ve taken great pride in our heritage as electronic artists,” Vaidya said. “And to be a great electronic artist, you need to have access to the best technology and tools. This has been a core principle of how we’ve been thinking about technology. We’re always looking for the best technology and solutions for creators. Over the last few years, we’ve been exploring, naturally, what are the possibilities to make available to people in terms of tools and technology?”
Vaidya said that while multiple AI technology companies were being considered and EA is in ongoing discussions for further opportunities, the Stability AI deal was closed because it could offer something no other company could.
“What stood out to us throughout the process, and why we ultimately decided to partner with Stability, was one of the things we talked about with all of our developers: the idea that you can’t control the tools that are available to you,” Vaidya said. “There are certainly gaps in capability, but even if the tools were capable, they’re not controllable. This is very frustrating for anyone to understand their usefulness in the context of their work. And then when you unpack it, why aren’t they controllable? It turns out that there are a lot of foundational models, common general foundational models, but they’re black boxes. Tools are very capable across a variety of tasks, But we’re not very good at it for very specific tasks. So how can we make it better for specific tasks? Customization means we need to be able to deploy the model, understand where the representations of different concepts and features are, and adapt them to specific needs at the same time. And we need a partner who can do that.
Regardless of how useful generative AI is to developers, it’s a scary addition to the industry, as many developers believe that as advances in the use of AI advance, their jobs will become easier but ultimately obsolete.
Stability AI CEO Prem Akkaraju counters this by bringing up the fallacy of economic lumps of labor.
“When you add or introduce technology to make something more efficient, you end up actually expanding your workforce,” Akaraju said. “The labor force is not fixed, so if it were fixed, as you say, there would really be no jobs.
Citing the example of the invention of ATM machines, Akaraju said, “When ATM machines were introduced in America in the ’80s, every bank teller in America was up in arms and said, ‘This is our job. We’re taking withdrawals, we’re taking deposits, and you’re taking our jobs away with this machine.’ And if you look from point to point, today’s bank teller jobs are 80. Why? Because the bank did the analysis and went from needing 30 people at each location to 13, and they did the obvious thing and expanded. Now there’s a bank on every corner. And because the bankers’ skills have improved, the average salary has gone up.”
Mr Vaidya said he “absolutely” did not want to “trivialize” the “fear and conversation around this idea of replacement” because “there are real concerns”. However, he believes that EA’s plans to use generative AI will lead to a change in the nature of available work rather than a significant reduction in the workforce.
“AI can play a role in enhancing certain tasks. Naturally, certain tasks will need to be reconfigured,” Vaidya says. “I think we need to get to that level to truly assess the impact on the workforce. Over time, new jobs will naturally emerge as well. And especially in the creative industries, we’re actually seeing more creators become part of the industry, because the reality is that entertainment is never a zero-sum game. It’s a growing pie. The demand for entertainment has been almost insatiable since time immemorial. And I think if these tools and technologies allow us to create and increase speed, quality, quantity, and scope, more creative collectives will be supported as demand increases. However, the shape of that population is likely to be different because AI will have an impact at the task level, resulting in a slight reconfiguration of the collection of tasks for any job. That’s not necessarily the case overall.
“I think the most important thing in a case like this is to be at the forefront and lead the conversation. Otherwise, you end up with this notion of disruption because you’re not ready for that change. And that’s especially true when that disruption comes from people outside the industry rather than from industry insiders. And that’s one of my biggest concerns. One is that we end up being observers, as opposed to participants and leaders in this conversation, because that’s what market forces do.” And today, there are many players outside of the creative industry who are looking to leverage AI in sports and entertainment. ”
