It, Robot

According to a recent article in the Economist, humanoid robots are getting close to reality. They can now fix you with a stare, cameras for eyes, trying to recognise faces and deciding who’s paying attention or making eye contact during conversations. 

By the way, I’m feeling a bit lazy today, so the informative, well-written bits in this blog are lifted almost verbatim from the Economist - ta very much hope you don't mind the plagiarism - and the flippant waffle is mine.

Roboticists (which sounds like a skin complaint) are trying to teach their robots manners, even though that would make them very unhuman: “Currently, it’s the worst ever party guest,” says one creator. “It butts in on every conversation and never shuts up.” 

Yup. I know plenty of humans like that and, after a few gins, I’m sure I’m guilty of a few monologues of my own. Talking about ill-mannered, Elon Musk, the new boss (scourge?) of Twitter, has unveiled his own robot, “a clunky, faceless prototype that walked hesitantly on stage and waved to the crowd”. 

Sounds like a hybrid of Keir Starmer and Joe Biden.

The reason for building a humanoid machine, apparently, is to perform tasks that involve human interaction. With a bit of development they might, for example, make a good companion for an elderly person – watching over them, telling them their favourite programme is about to appear on television and never getting bored with having to make repeated reminders to the forgetful. Please can I have one to keep an eye on Hubby? They could also learn to play board games, but only well enough so that they remain fallible and can be beaten. That is soooo condescending!

To interact successfully with real people, a robot needs a face: “The human face is the highest bandwidth communication tool we have,” observes Will Jackson of Engineered Arts, a small robotics company in Falmouth. “You can say more with an expression than you can with your voice.”

This is true. I’ve been known to smile with my mouth, look daggers with my eyes, and you really have to watch out when my nose twitches, like Samantha in Bewitched (I’m showing my age). 

While it’s possible to design human-looking robots, some are designed deliberately to look how people might expect a robot to appear, e.g. pore-less complexion, visible joints and no hair. Are we talking Joan Rivers or Matt Lucas here? People, we are told, can be unnerved when an artificially created being shifts from looking clearly not human into something more real, but not quite real enough. Comfort levels rise again as similarity to a human becomes almost perfect. This means we need something that looks like a robot, or like a human, though not something that looks not quite like a human. That rules out Jeremy Hunt then.

Some roboticists do, however, seek visual perfection. Ishiguro Hiroshi, director of the Intelligent Robotics Laboratory at Osaka University, recently unveiled a robot that resembles Kono Taro, Japan’s digital minister, through which he can speak with his own voice. This robot could be used to stand in for the minister at some functions, like the robots currently deputising for Diane Abbott, although someone needs to repair her frequent short-circuiting.

Shadow Robot, a firm in London that makes one of the most dexterous human-like robot hands available, traces its roots to hobbyists meeting in the attic of its founder’s home. Hobbyists? Not the word I would have used. Have you ever seen the episode of the Big Bang Theory where Howard has to be taken to the ER because, errrr … watch it on YouTube, coz I'm saying nowt else!

For robots to interact safely and successfully with people, we need safety systems. At present, the use of robots is governed mainly by standard safety and product liability rules. Some argue that special robot-specific laws will be required. As every sci-fi buff knows, Isaac Asimov laid out a set of these eighty years ago that were regurgitated for the Will Smith film, I Robot. They are:

• A robot may not injure a human being or, through inaction, allow a human being to come to harm

• A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law

• A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

Buffs also know that SF storylines often revolve around these laws not quite working as planned. I, Robot being a case in point: ‘Sonny’ was programmed by his ‘father’ to break the first law for, in a nutshell, The Greater Good.

In one respect I love the idea of having a robot help me round the house and run errands – 100% obedience, behaviour and successful outcomes, on time and under budget: think the exact opposite of Hubby or HS2. But having him / her / it (wokeism has invaded robotics) analyse me, misunderstand me then judge me? Gotta get me a BoJo-RoBo to make sure that doesn’t happen, or I shall have to invoke my own three laws:

• I may injure a robot or, through inaction, allow a robot to come to harm if it pisses me off

• A robot must obey the orders given to it by me as I intended, even if I’m unclear or contradictory, the first and third laws kicking in otherwise

• A robot must self-destruct in 30 seconds if it pisses me off, then clean up its own mess to save me the bother.

It’s not as if any such upcoming invocation of mine would be a surprise to a humanoid robot. It should be able to register and interpret my sweet smile, dagger-eyes and twitching nose.

I wonder how they might react to me on my broomstick.