Public access radio that connects community members to one another and the world
Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations
KDNK is hiring a Membership and Events Director! Click here for details

The hit horror movie 'M3gan' taps into our fears around artificial intelligence

ARI SHAPIRO, HOST:

The hit horror movie "M3gan" imagines a humanlike robot doll that goes rogue.

(SOUNDBITE OF FILM, "M3GAN")

JENNA DAVIS: (As M3gan) Don't worry, Cady. I won't let anything harm you ever again.

SHAPIRO: It speaks to our growing fascination and anxiety over artificial intelligence. In recent months, AI chatbots and image generators have captured everyone's attention. For our weekly All Tech Considered segment, NPR's Bobby Allyn is here to help separate science fiction from something actually close to reality. Hey, Bobby.

BOBBY ALLYN, BYLINE: Hey, Ari.

SHAPIRO: For anyone who is not familiar, "M3gan" involves a very old film trope, right? It's the robot that turns on its creator.

ALLYN: That's right. The film is about a toy company that makes a lifelike AI doll named M3gan. It becomes the best friend of a young girl whose parents died in a car crash. And the doll is pitched as a companion for the girl, and it's programmed to comfort her. But then M3gan goes haywire and starts killing people.

SHAPIRO: And why, apart from the dancing, are people so into this movie?

ALLYN: Yeah, it's causing a stir because I think it portrays our worst nightmare about AI and robots. And, you know, what's different this time compared to, say, robot films of 10, 15 years ago is that the idea of a humanoid robot was so much more farfetched back then. And now it almost feels like it wouldn't be crazy if a big tech company announced a robotic doll companion like M3gan, say, next year. People wouldn't be too shocked, I feel like.

SHAPIRO: Oh, yeah. I watched those Boston Dynamics videos of robots doing crazy things, and it, like, gives me chills. AI has actually made so many strides in recent years. Tell us where it stands now.

ALLYN: Yeah, so we're nowhere near robot sentience, but, yeah, let's run through some examples, like the Boston Dynamics that you just mentioned. So Elon Musk brought out a humanoid robot to the stage of a Tesla event that did a raise-the-roof dance that was a little weird and creepy. There's an AI tool out there that can actually listen to court arguments and then helps defendants fight traffic tickets in real time - so in front of a real judge. And yes, then there's a company, Boston Dynamics, and their videos have really made the rounds. You know, it shows these robots just doing obstacles and doing backflips and coordinating with each other. I actually showed it to a friend of mine, Ari, and I said, take a look at this. And she goes, that's CGI, right? And I said, uh, no, that's not film animation. This is literally the state of robotics right now.

SHAPIRO: OK. But have there actually been any killer dolls yet (laughter)?

ALLYN: No.

SHAPIRO: So why are we also spooked by "M3gan?"

ALLYN: Yeah. I think, you know, AI technology has been developing for decades, but it feels like it's gotten really good, basically, overnight because lots of different examples are hitting the mainstream at once, whether it's ChatGPT or Lensa. And we're all just very dazzled by these, right? But AI scholars are really emphasizing right now that machines cannot think and reason like a human, let alone turn on their creators. Sorry, this does sound ridiculous to say it out loud, but AI bots do not have common sense, right? They don't have the ability to think. They're a program by engineers. They run on code to complete tasks or to look for patterns and then mimic those patterns.

SHAPIRO: OK, that's what you say now. But seriously, is this PSA necessary just because of a movie about a killer girl doll who wears sunglasses?

ALLYN: I think so. I mean, "M3gan" has become this jumping-off point for actually important and serious debates around the ethics of AI - around questions like, what should the rules and laws be around privacy? Who is liable when an AI robot destroys or hurts someone? I mean, who owns the images and text that these AI bots are creating from the entire knowledge of the internet? These are complicated and thorny things. And Ari, we can thank "M3gan" for giving these issues more oxygen right now.

SHAPIRO: NPR's Bobby Allyn, thanks.

ALLYN: Thank you.

(SOUNDBITE OF MUSIC) Transcript provided by NPR, Copyright NPR.

Bobby Allyn is a business reporter at NPR based in San Francisco. He covers technology and how Silicon Valley's largest companies are transforming how we live and reshaping society.