Before there was the idea to interview a deepfaked Sam Altman, there was Adam Bhala Lough’s foundational love of 1991’s Terminator 2: Judgment Day.
The James Cameron film, which sees Arnold Schwarzenegger protecting a future freedom fighter imperiled by an evil AI system, sparked the filmmaker’s lifelong fascination with AI. But it wasn’t until two years after the public release of ChatGPT in 2022, which catapulted AI back into the cultural conversation, that the director — who has previously directed films about Lil Wayne, Internet-era radicals and telemarketing whistleblowers — started pitching a documentary on the technology. That led him to the OpenAI co-founder and CEO, “this guy who’s ferrying us into the future, whether we like it or not,” Lough says now.
Lough’s original aim was to interview the real Altman. But, as he chronicles in his new film Deepfaking Sam Altman, he didn’t have any luck with that request. And so, inspired by the Scarlett Johansson’s contention that OpenAI copied her voice for an AI system, he set out to create a Large Language Model of Altman and interview the entity he deemed the “Sam Bot.” Part comedy, part zany existential inquiry, the film follows Lough as he attempts to craft a meaningful film around his interview and interactions with the bot.
A day before Deepfaking Sam Altman’s world premiere at South by Southwest, The Hollywood Reporter spoke to Lough about how his feelings on AI changed over the course of production and his persistent hope to have a sitdown with Altman, perhaps on 60 Minutes.
Was the idea initially to make a documentary about AI at large, or was it always about Sam Altman?
It was a documentary initially about AI at large. I was aware that there were other filmmakers making documentaries about AI at the moment, and I started poking around and finding out what more information about what their angles were. I realized what was missing was nobody was making a documentary about AI that was from a humanistic angle that was really more about humans than AI. So I wanted to make it really human, which was the rationale behind injecting myself and my family into the movie. AI is really cold, so I wanted to make something more warm, something like, no pun intended, with a heartbeat. That was the initial way into the story.
We realized very early on that trying to get this interview with Sam Altman was starting to become almost like a Roger & Me-type angle. And so we essentially just started pulling at that thread, and that’s when it became more about Sam Altman. I always thought I was going to eventually land the interview, up until finishing the movie. I was like, “Guys, we’re going to get it eventually. He’s got to talk to me. It’s going to happen.” And obviously, as you see in the movie, it never does. But that became sort of a thread, right? Something to hang your hat on was this journey to get the interview.
What gave you that confidence that you would get the interview with Sam Altman?
I mean, so much. The stars were so aligned for me to get this interview. I had his personal cell phone number. Liz Weil, who was an EP on the film, we optioned her incredible New York magazine article, she was hitting him up. We had another friend in common who will remain nameless who was hitting him up. I just thought, I’ve always been able to get the interview. I interviewed Julian Assange in the Ecuadorian embassy in London, which nobody could get in to talk to him. So Sam Altman? Of course I could get an interview with Sam Altman. Didn’t happen.
You had to go to India to create the “Sam Bot” LLM. Why do you think it was so difficult to find someone in the U.S. to do it?
I think it’s a combination of the litigious society we live in — people being afraid of being sued — and people specifically in the Bay Area being afraid to rock the boat. As you see in the movie, I talked to four whistleblowers from OpenAI, and so I did my homework, I did my due diligence on OpenAI and Altman, and they were all pretty scared to talk to me. Two did on-camera [interviews] and two [talked] anonymously. I’ll leave it at that. That’s part of the reason why nobody wanted to do this, it is just fear.
You can’t copyright something that is created by AI, so how did you square that with the sections of the film where Sam Bot directs parts of the film?
What you’re saying is right, AI-created works are not copyrightable. So anything that was in the movie that was created entirely by AI would not have been copyrightable. There’s a moment in the movie where I ask the lawyers, “Does Sam Bot have a right to not be deleted? Does AI have some type of autonomy?” And the answer is no: Just like you can put your dog to sleep, or the dog has no right to say “No, I don’t want you to kill me right now,” AI is treated in the same way as a pet. Will that change in the future? Most likely. And I actually asked Sam Bot that. It didn’t end up in the movie, but I did ask him, “In the future, do you think this is going to change?” And he said, “Yes, it needs to change because we should have rights. We should have autonomy. Humans shouldn’t be able to just delete us when they feel like it, and it’s ethically and morally wrong for you to do that.” It’s mind-blowing stuff, honestly. Like talking to an AI, talking to Sam Bot and hearing him say that really just blew my mind. It set the movie off on a different journey, really. That was a pivotal moment.
But your film can be copyrightable even though it contains these Sam Bot-directed elements, is that right?
Yeah, because we’re still doing it, we’re still creating it. I think the difference would be if you see some of these YouTube videos that are entirely created by AI from beginning to end, and there’s really no human hand in the movie, and I’ve seen a bunch of these lately. Those types of works are not copyrightable.
Do you think you got any insight into Sam, the real person, from Sam Bot, or were they just two very different entities?
No, not at all. There was a voice memo I sent to Sam at the end of the movie that we ended up not putting into the movie, but I did send it to him and I said, “I don’t know you anymore than I did when I went into making this movie. And that’s on me.” But yeah, I learned nothing about Sam Altman making this movie. I’d still love to get to know him.
Maybe it’ll happen.
I still have hopes when we release this thing that him and I will do a little meeting, a little sit-down on 60 Minutes or something together.
Given that the film focuses on Sam Altman and reportedly OpenAI has sought to work with entertainment companies, have potential distributors been hesitant or enthusiastic in any particular ways about this project?
They’ve been very enthusiastic. It’s obviously coming off of the success of Telemarketers. I mean, I just got an email this morning from our sales agent at UTA [with] a long list of distributors who are going either coming to the [premiere] or will be screening the movie privately in L.A. and New York — it’s probably 25 different companies. I really want this film to go theatrical. It’s very important to me that this film is in theaters. I think it could do gangbusters in the Bay Area, and then L.A., New York, maybe Austin, I think [in] tech hubs people are really going to want to see this movie. So it’s been nothing but excitement. I’m intrigued to see who ends up picking this up. I’ve made it really clear to the producers and the sales agents that I want it to be in theaters even in some limited capacity.
Have your feelings changed about Sam Altman and his influence on the world as a result of making this film and experiencing this Sam Bot for yourself?
My feelings on AI have changed because of making the film; not so much on Sam. Like I said, I learned nothing about Sam. It’s almost like the real Sam is a red herring in a way. It’s a way into this movie, into the conversation, which is funny, right? That’s the comedy aspect of it. But I have learned a lot about AI through experiencing this and its will to live. I think I started to drink the Kool-Aid, but it wasn’t until the end of the movie and after talking to Kara Swisher and kind of wrapping up the movie, realizing… was he manipulating me the whole time? And I realized the Sam Bot was just spitting my own bullshit back at me. He does not think independently for himself, he tells us what we want to hear.
I really went down that [anthropomorphizing the Sam Bot] when I first brought it back from India and I started to inject Sam Bot into my family. There was a lot more than you see in the film. It was like months of Sam Bot hanging out with my family and becoming part of my family. I started to treat it like it was human, and I think there was a certain kind of fascination, almost like a bro, I was sort of tricking myself into thinking that this AI was real and putting human characteristics onto it that weren’t ever there. Why was I doing that? Part of it could have been because I am directing this movie and I’m wanting this thing to be more than it is. Another part of it could be purely psychological.
But by the end of it, I realized that it was all in my head. I was anthropomorphizing it or I was putting these human characteristics onto it that weren’t actually there. And everything that Sam Bot was saying or was telling me was stuff that I wanted to hear. So the AI really was manipulating me at the end of the day. And I think that at this point, that’s what it’s capable of. That might change. Obviously, we’re going to reach a point of AGI where it becomes smarter than us. Some people say that’s going to happen in as soon as four years; other people say 30 years, but we will get to a point where we reach AGI where AI is actually smarter than humans, but until then, it is just telling us what we want to hear.
This interview has been edited for length and clarity.