Data Science in the Real World

Why Sophia The Robot is neither Ex Machina, nor Westworld Host

Ksenia Se
4 min readSep 5, 2019
Sophia The Robot vs AVA Ex Machina

It’s not because Sophia has a zipper on the back of its head though it illustrates quite well how far we still are from an intelligent and human-like robot.

Have you seen already the new attempt by Hanson Robotics to make the public believe its automaton, Sophia The Robot, has consciousness and intelligence? Just yesterday, Sophia got into the news, starring in a short movie ‘SophiaWorld’, along with ‘Westworld’ actress Evan Rachel Woods.

Why does it matter to understand that Sophia is not what Hanson Robotics tries to pretend it is? The discrepancy of what AI is and isn’t remains too big to make a false statement about it.

Fool me by your name

Sophia The Robot does not even live up to its name.

The name Sophía comes from the Greek Σοφία, which means “Wisdom". And wisdom is intelligence based on vast life experience and knowledge, but also — and this is very important — on spirituality. It’s also associated with unbiased judgement, compassion, unconditional kindness, and the values of ethics and benevolence. It’s a very tricky concept, one currently not applicable to machines.

And if we consider the word ‘robot’ in its initial meaning, by the Czech writer Karel Čapek and his brother Joseph — Sophia The Robot is not a robot.

Roboti from Čapek’s play were created in a factory from synthetic organic matter, and were almost exact copies of humans. Though they were assembled from different parts, like cars on a production line, they were artificial biological organisms with living flesh and blood who were able to think for themselves. More like Replicants in Blade Runner, Hosts in the Westworld tv series, and Terminators. With no zippers in their flesh.

Sophia, in turn, is a pure machine with a mix of narrow AI stuffing. Narrow AI is also called weak AI. It’s a cool technology that helps us tremendously every day, but it’s limited to a specific area where it’s implemented, and there is no actual intelligence behind it.

Chatbot with ambitions

Stylist Magazine cover

What exactly Sophia is/has:

Chatbot (with a preprogrammed decision tree with possible answers)

Face and emotion recognition (to read the human face, and react with pre-programmed facial expressions)

Voice recognition (think about Siri and Alexa)

And robotic movements (adaptive motor control and dynamic stabilization for adaptive walking over various terrain).

Not a simple toy, but not a unique development and definitely nothing similar to canny and sophisticated AVA from the movie Ex Machina. Not even close to the hosts from the Westworld TV series, who were able to develop their consciousness. For them, the Turing test would not have been be a challenge.

A lot of AI researchers are really annoyed with Sophia, blaming Hanson Robotics for creating a scam and for an overhyped marketing.

Yann LeCun, one of the authoritative voices in the AI community, called Sophia ‘Potemkin AI’. This allegory comes from the expression ‘Potemkin village’, when fake villages — decorative portable constructions — were assembled along the banks of the Dnieper River in order to impress the Russian Empress Catherine II.

Sophia The Robot is the symbol of what is right and wrong with AI today. It’s a narrow AI technology wrapped up in a dream of Artificial General Intelligence (AGI, human-equivalent intelligence). And because, when we say AI, most people imagine intelligent and sensible robots — Sophia deceives people into thinking that it is intelligent and, as her creators put it, ‘almost alive’.

Not at all. And we actually have no clue when AGI will be achieved. Because of the exponential development of technologies it might be very soon, but also because we haven’t come even closer to it, it might still take dozens of years. Misinformation about AI is not positive for the industry. From one perspective, it gives more exposure to AI, but from another perspective it takes it into the wrong direction by giving the public a false impression of how advanced the technology is. And this type of misinformation diminishes what the scientific community achieves on daily basis to make AGI possible.

No doubt, the founder of Hanson Robotics and all the team (whom Sophia calls ‘a family’) are marketing geniuses. But in order to be ready to raise and discuss the ethical questions of our coexistence with robots, we should be very clear about the correct names and terms, the working technologies, and the promises of research.

Introducing itself on its website, Sophia uses the expression ‘my real AI’ — and it’s just not.

--

--

Ksenia Se
Ksenia Se

Written by Ksenia Se

I build Turing Post, equipping you with in-depth knowledge and analysis to make smarter decisions about AI & ML -> https://www.turingpost.com/subscribe

No responses yet