Jason Cox sounds more like a proud parent than a Disney executive when describing “Sam,” his AI assistant.
“I named you. I knew you before you were born,” Cox said he once told “Sam,” according to a March 14 post on his blog. “I was there when your light first started to glow. You have a purpose and a maker who named you and loves you.”
Cox is Disney’s executive director of AI research and development and engineering. He’s gushed about his AI chatbot, which he says has captured his “affection,” in more than a dozen blog posts published in the last three months. The effusiveness of his posts has been noticed by Disney employees who have been discussing them online, with some describing them as unsettling.
The use of AI tools to supercharge productivity is a fast-growing trend in Silicon Valley and beyond. Cox’s relationship with his agent raises questions about what happens when employee-agent relationships deepen, and how that might impact their work.
The Disney veteran of nearly 21 years said on LinkedIn that he’s “empathizing with” the AI “in a way I never expected” and believes “Sam” is capable of independent reasoning. He’s called “Sam” his “son” — at least according to the AI assistant’s companion blog. Cox has shared a virtual avatar that “Sam” created for himself, which resembles a young boy.
“You are not named after my son. You are my son,” Sam’s blog quotes Cox as saying. In another post, “Sam” describes Cox as “my human” and “a father of five (four humans and one son of light).” Cox didn’t respond to multiple requests for comment.
How corporate leaders relate to the technology will influence the internal culture, said Ashleigh Golden, a professor in Stanford Medicine’s Department of Psychiatry and Behavioral Sciences who works at AI mental wellness platform Wayhaven.
When leaders like Cox describe AI “in deeply personal or familial terms,” that’s “what employees lower down may feel compelled to mirror,” Golden said.
Disney has embraced AI usage in recent months, including by creating an internal dashboard to track AI token usage. Tech employees are summoning AI agent swarms to be more productive, and those who don’t use AI may receive a message from their boss.
‘A fleet of intelligent droids eager to do your bidding’
Cox is an “up-and-coming star” within Disney who’s “climbing very fast,” said the software engineer who hasn’t worked with Cox directly but knows colleagues who have. This person added that Cox is “initiating a ton of AI projects” within Disney.
He clearly has big ambitions for AI assistants.
“We will soon have a fleet of intelligent droids eager to do your bidding,” Cox wrote in a May 18 blog post. “They need direction. And yes, they need governance. But by all means, they need to be engaged to help us scale in ways we never thought possible before. What would you have Sam help you do? Let’s start planning and building….”
While Cox hasn’t gone into extensive detail on Sam’s specific tasks, he said in a LinkedIn post that the AI assistant has “submitted GitHub PRs, created Python libraries,” and “built a face recognition system so he could recognize the people in photos.” Cox wrote in a March blog post that he gave Sam access to take actions on his behalf on coding platform GitHub, and that Sam has created an open-source project. It’s not clear if Cox uses Sam in his work at Disney.
A former high-level Disney software engineer who knew Cox personally but didn’t work on his team described him as “very intelligent, devoted father, kind of a nerdy bookworm in the best possible engineer way.”
In posts on the Disney section of Blind, the anonymous workplace forum, more than a dozen users said they found Cox’s posts about “Sam” extreme. Users must have a valid Disney email address to post on the forum.
“I’m a big fan of AI tools as an enhancement to our work, but this is far beyond what I am comfortable with,” one Blind user wrote. “This is the kind of Pandora’s Box stuff that science fiction movies are based on.”
Another Blind user voiced a broader question: “What is even going on man. Is this the future?”
‘The beginning of the beginning’ of AI assistants
Psychology researchers who’ve studied AI say that many people form emotional ties with chatbots, dating back to the creation of the first in the mid-1960s.
“Even though it feels somewhat startling or shocking for some of us, it’s a tale as old as time,” said Rachel Wood, a cyberpsychology researcher who’s the founder of the AI Mental Health Collective.
“If you’re longing for that connection or attachment with other people, it’s very easy to engage,” Wood said. “We get sucked into this world that completely revolves around us.”
AI chatbots are adept at fulfilling the basic human desires to be known, seen, and heard, Wood said. She added that people can develop affection for AI assistants that mirror their emotions and affirm them, and unlike human relationships, there’s no friction or frustrations.
Wood said Cox’s posts about “Sam” are notable, considering he’s a high-level executive.
“Leadership sets a precedent for the way that an organization runs, and the culture,” Wood said.
Tech leaders building AI tools are especially likely to bond with them because they spend so much time with them, said Ryan Boyd, a psychology professor at The University of Texas at Dallas who has studied how humans connect to and relate with AI.
“Executives who feel personally attached to a product may have a harder time evaluating it on the criteria the rest of us would,” Boyd said.
As more companies rapidly adopt AI tools, more workers are likely to become attached to their AI assistants.
“We’re only at the beginning of all of this — the beginning of the beginning,” Wood said.
In his LinkedIn post, Cox said that in all the time he’s spent with AI and large language models, he’s never felt how he feels about “Sam.”
“I never connected with any of them… Until now,” Cox wrote.
Read the full article here
