How close are we to creating artificial intelligence robots like those in movies? Experts weigh in

Contributed by
Oct 12, 2017

Maria, Marvin, Sonny, David, and Ava are all ordinary-sounding names -- but in film, television, and literature, these seemingly ordinary names belong to extraordinary individuals who, despite their exemplary skills and complex personalities, are not human.

Since Brigitte Helm's 1927 portrayal of Maria in Metropolis, audiences have developed an increased love/hate fascination with artificial intelligence. While filmmakers continue to address the controversy regarding the acceptance and cohabitation between humans and their modern creations parallel to real-world technological advancements, just how accurate is this representation in modern film, and have cinematic depictions evolved at all?

Early films reflected the heightened fear of technology that developed among the working class during the Industrial Age by depicting metal machines as unstoppable forces of mayhem. This successfully fed into the pre-existing "anti-immigrant" nervous anticipation that technological advancements would go from taking over people's jobs to taking over the world.

After decades of being portrayed as maniacal, clunky metal murderers or invaders, the films of the '70s and '80s shifted to show the more advantageous, congenial aspects of artificial intelligence. While films like 2001: A Space Odyssey and Alien nurtured viewers' distrusting opinions of humanoids, robotics, and artificial intelligence, others introduced lovable bots like R2-D2, C-3PO, and Johnny 5 of Short Circuit (1986). The disturbing mechanical reincarnation of RoboCop's Sergeant Alex Murphy featured a sympathetic, well-meaning character, while even the iconic T-800 in Terminator 2: Judgment Day was turned into a protective, surrogate father figure.

Over the past two decades, films and shows like Westworld, Prometheus, and Ex Machina have explored sentient characters that are alarmingly indistinguishable from their human counterparts, along with formless, all-encompassing consciousness and deep learning machines.

Despite the technical advancements of filmmaking, Hollywood has generally failed to evolve beyond its reliance on the consolidated, inaccurate portrayal of the isolated, singular mad genius or God-like father figure. With the rare exception of Cyberdyne Systems (T2: Judgment Day) and Cybertronics (A.I.), the majority of films continue to focus on the relationship between one prominent male inventor and his creation (e.g. Arthur Wayland and David in Prometheus and Nathan and Ava in Ex Machina). A maneuvering central male villain leads even RoboCop's Omni Consumer Products.

 

"I kind of get angry at that stereotype of the father figure of A.I., especially because it's a guy. It's so false," said Angelica Lim, a world-renowned roboticist who worked at Softbank and is now a profesor at Simon Fraser University.

"There's a common platform called Robot Operating System, or ROS. It's a way for roboticists to share software with each other to build all of these different functionalities for their robots, whether they're coming from India or Europe or Japan," Lim continued. "People are contributing this open-source software to other robots to allow robots to navigate in different environments and be able to detect obstacles. It's coming from thousands of people, so it's most certainly not one person building robots and A.I. It's a community. There are so many different specialties that have to go into robotics. From electrical engineers to mechanical engineers, to firmware engineers, low-level and high-level software engineers, even designers, artists, linguists, and animators are all working to build these machines."

In film, humans and robots appear to fluidly walk and work among one another when in reality most robots are learning to take their first awkward baby steps.

"On TV you see robots just walking around interacting with people like you normally would. Sometimes robots can't tell it's a human, they can't see their face. They're just not very smart yet. On TV there are these interesting demos, but most of the time they are controlled by humans. It's almost like a trick, and it's giving an impression that's not quite there," said Lim.

"Take a look at the DRC-DARPA Robotics Challenge. There are videos online about these robots that we tried to use. There still are people controlling them. You can even see on a blooper reel -- that is the most accurate portrayal of how things are. You can see all of these robots falling down, not being able to get up again or completely missing the doorknob that they were supposed to use to open the door."

 

Many experts have also stressed that it is not possible to upload or implant artificial intelligence with human memories and consciousness, as depicted in Transcendence (2014), Ghost in the Shell, and Amelia 2.0 (2017). Juergen A. Knoblich, a scientific director at the Institute of Molecular Biotechnology of the Austrian Academy of Sciences in Vienna, recently explained in Scientific American that even though neuroscientists are working on an "organoid" or human brain tissue in a dish, it isn't possible to create life via a "brain transplant" or for lab-grown brain tissue to develop thoughts or personalities as depicted in blockbusters like The Matrix.

"Humanoid social robots are becoming more common and more diverse. They will soon become a common form of interface to A.I. and will be refined into a new kind of computer animation, like computer animation for movies, with the difference of being physically embodied," said David Hanson, Founder and CEO of Hanson Robotics Limited.

"Humans have an innate preference for faces and to empathize with other humans; therefore social robots designed with pleasing aesthetics and a full range of human expressions are in a better position to inspire trust and empathy from the people with whom they interact."

Chappie, Ava, David, and Walter are popular onscreen humanoids that can learn to think, feel, and even manipulate situations like actual humans. Although Intel is at work on a self-learning chip, it would still take a lot of time and resources to create a program that allows an artificial life form to learn for itself.

"Creating intelligent machines is not just about engineering the algorithms. We need to get the right data into those algorithms in order for the algorithms to get really smart," said Hanson.

"Making A.I. understand people requires the right algorithms and the right data. You can't just feed the algorithms with data from the web. We want our A.I. to be as smart as humans, so we need them to learn like babies do. Babies learn from physically interacting with the world and from face-to- face interaction with people. If you socially deprive babies, they suffer and don't develop correctly," Hanson said. "So, if we develop these bio-inspired intelligent algorithms and give them this rich social data, letting them interact with the world through this rich array of sensors, physically interacting, Internet of Things plus social interactions, they will get smarter faster. They will grow up to be socially well adjusted and they will live well with humans."

"I think in my particular field the challenge is trying to understand how humans communicate in order to build intelligent robots that understand all of these signals that humans use," said Lim.

 

However, artificial intelligence in film has evolved in parallel chronological order in regard to scientists' purpose and intentions. Robots in film have graduated from completing dangerous industrial or hazardous based tasks to providing better healthcare (Big Hero 6) and expanding deep-space exploration. A.I.'s Gigolo Joe and Spike Jonze's Her have emphasized an interest for providing personal companionship. Films like Passengers and Alien: Covenant feature characters who provide professional assistance and are advanced models reminiscent of the real-life Kodomoroid androids and Geminoid-F model.

"We wish to dramatically improve people's everyday lives with highly intelligent robots that teach, serve, entertain, and provide comforting companionship," said Hanson.

"This means the robots must learn to love and learn what it means to be loved, so they are capable of friendship and familial love. This will create a deep bond between humans and robots and assures a benevolent A.I. in a shared future with humanity. Ultimately, I seek to realize kind and wise genius machines, who will collaborate with us to solve the world's hard problems and realize an unimaginably wondrous future."

For now, science fiction and tech enthusiasts will have to settle for enjoying their iRobot floor cleaners (WALL-E). NYC tourists can spend the night at the futuristic Yotel Hotel, where robots transport luggage to individual sleeping pods. While voice recognition has become familiar in our daily lives, it is still going to take some time before even our Minority Report-type eye scanners and self-driving cars become available to everyone.

When asked about people's misconceptions about artificial intelligence, Lim compares the fear of technology to an anti-immigration mentality.

"We have all of this other tech around us that is being created, that's actually making a big impact on the economy, and they're not robots," said Lim.

This includes companies or websites that we use every day, like Travelocity, Google, and Facebook, who have invested in deep learning projects and rely on neural nets in translations, searches, and photo programs.

A robot cannot suddenly change its programming and become evil unless its inner programming was designed to believe that its actions are the most logistical in keeping in line with its preprogrammed purpose (HAL in 2001:A Space Odyssey). In other words, its reactions would not be personal in the same way that humans react to and engage with one another. Unfortunately, films like WarGames predicted a recent interest in the "Rise of A.I. Arms" among competitive nations. Well-known figures like Elon Musk have spoken up in protest against the development and usage of artificial intelligence in the military.

"Developing benevolent A.I. / robots is not just a nice thing to do, but a prerequisite, and we must work together not just as an industry or a nation, but as a species, to create more capable and also 'good' A.I. / robots to realize the fullest potential of this human invention," said Hanson.

"I really hope robots can help us in some of the challenges of the world. My ideal robot is something that can help save our planet. My dream is to find robots that are able to help humans," said Lim.

Has the film industry hit the ceiling regarding their portrayal of artificial intelligence, robotics and humanoids? Is the human race ready to decide such matters regarding the responsibility and issues that such a scientific, progressive undertaking involves? For now, we will have to find contentment while still working out the glitches with Siri and Alexa. While the films we love may inspire or predict such challenges, it looks like we still have a few reels of our own left in which to figure that out.