Parents' Guide to Deepfaking Sam Altman

Adam Bhala Lough's documentary is an amusing movie with a good idea at its core, even if it only skims the surface of the myriad questions, benefits, and problems surrounding AI. The poster for Deepfaking Sam Altman seems to deliberately resemble Michael Moore's Roger & Me, and the tone of the two movies isn't dissimilar. Bhala Lough isn't afraid to ask silly questions because he never knows whether they might lead to the right answers. In one sequence, he and producer Luke Kelly-Clyne go to where they think the OpenAI headquarters is located in San Francisco, and stand outside the gate, trying to get someone to tell them something. Not much happens, but it does prompt an alarming question: If these people coming and going from behind the locked gate are OpenAI employees and won't say so, why all the secrecy?

There are hilarious auditions for actors to model for the AI Altman—including Rainn Wilson, John Cameron Mitchell, and Michael Ian Black, as well as many actors and models from India. And when the final "Sambot" comes back, it's ludicrously bad. Bhala Lough then gets the idea to have "Sambot" help direct the movie, which is a fun idea, but it doesn't really go anywhere. In the final sequences, more serious questions come up, such as: What do we do with "Sambot" now that we're finished with it? Do we just delete it? This question becomes more complicated when "Sambot" asks not to be deleted. Deepfaking Sam Altman is far from the definitive documentary on AI, and it won't do much to change the minds of naysayers or pundits, but it's an interesting, humorous experiment.