Guild Wars 2

Brief Summary
For those who don’t know this robot first appeared as Job-o-Tron during the Secrets of Southsun release when he was assigned to ‘greet’ settlers as they arrived, kitten their skills and assign them jobs. Which happened to involve a lot of insulting them for not already being employed and/or not having suitable skills.

When the Consortium lost their hold over the refugees he lost his job and became Hobo-Tron. He next appeared in Divinities Reach as a busker, was captured by Scarlet and later freed, then arrested and charged with helping her.

He’s re-appeared again as Ho-Ho-Tron in Lion’s Arch – forced to raise donations for the victims of Scarlet’s attack as penance for his supposed part in her scheme.

There’s an interesting article here about how he was created and developed:
https://www.guildwars2.com/en/news/job-o-trons-journey-the-evolution-of-a-gag/

Summary Ends

His story is interesting to me because in a way it’s very unusual. The impression I’d gotten before now is that golems were somewhere along the lines of modern computers and elementals. They’re magical constructs that can be programmed to retain and recite information and follow instructions. Sometimes quite complicated instructions but even so that’s ultimately all they’re doing.

Ho-Ho-Tron however has even expressed emotions and hopes and fears about the future and has done things which are almost certainly far outside his programming. In short he seems to be able to think for himself.

Whilst this would be amazing in real life where research into artificial intelligence (actual AI, not the programs used in games) is still struggling to come up with something that can do more than follow very complicated instructions it’s maybe not surprising in Tyria where there’s a lot more flexibility. Whose to say how magic works? Or a combination of magic and technology. Maybe most golems are deliberately made less intelligent than they could be because it’s more convenient for their owners.

What is surprising is that the people of Tyria seem to treat Ho-Ho-Tron as a sentient being rather than a machine. Normally if a golem went hay-wire and was caught assisting a criminal his owners and/or creator would be blamed for not having better control over him (as happened 250 years ago when Zinn was put on trial because his golems tried to kill some of the most important people in Tyria at the time.)

At best the golem would expect to be re-programmed, at worst shut down and scrapped. Instead he’s been treated like a living creature, literally, he seems to have gotten exactly the same treatment as Marcello (aka The Minstrel).

If you know anything about artificial intelligence/artificial life research, or in fact research into behaviour and intelligence in non-human species you’ll know this is one of the major stumbling blocks – trying to get anyone to concede that such a thing could exist, let alone already does. Research into intelligence in animals has almost always been defined as an attempt to answer the question ‘what sets us apart from animals?’ (with a lot of frantic shifting of goal posts as suggested answers are proven wrong) and it’s only recently that the perspective has begun to change, with a lot of opposition.

Maybe this is where the people of Tyria have an advantage – there are a huge number of undeniably intelligent, sentient races on that planet. Not just the 5 ‘main’ playable races but the centaurs, grawl, quaggan, kodan, hylek, even the ogres and many others. So maybe it’s easier for them to accept that any being can be sentient, even one they wouldn’t normally expect.

But I still find it surprising, especially in that no one seems to have questioned it. I wonder if I’ve missed some precedent (I admit asura are not my strong point when it comes to lore), or maybe this is a new step forward following things like the charr-human treaty and the Pact. It will definitely be interesting to see how it progresses, both in terms of what Ho-Ho-Tron does and how people react to him.

There is that conversation in Rata Sum between two elder asuras about some golem uprising some years/decades ago. Cant remember where it is though.

One of the asuras was hanged by his collar on a road sign for asking directions from a golem. Appearently they hate(d) that.

There is another conversation near the Rata Sum bank/trading post, someone asks a peacemaker what is preventing the golems to gain sentience and go hostile in an attempt to overthrow the asuras. Cant remember the proper terminology, but there is some kind of prohibitor i think that serves exactly that purpose.

On the other hand, golems seem to be very easy to make for asuras, seeing how some or most pre-collage progenies can build their own. Some might not include such a prohibitor in its creation, and that can go in any direction plotwise.

Come to think of it, its a miracle that there havent been AI disasters where a progenies ingenious but dangerous little project goes haywire and starts reproducing itself, just to top it off.

They had golem insurrection. It’s why golems have inhibitors now.

Maybe after he lost his job, the programming went awry and he overcame his inhibitor? It seems the real shift is after Southsun’s events, to me the article claims that it’s just a regular golem made for a task of slave driving.

The Job-O-Tron is likely ownerless by now since the engagement of its “Tertiary directive”, hence why the golem itself was put on trial rather than its former owner of whom the Human ministers may not even know of.

Rather than free, independant thinking, Ho-Ho-Tron’s current acts may be nothing else than adapting to calculated odds for its prolonged existance – what’s this Tertiary objective?

“Job–o–Tron: Departure–of–the–settlers–diminishes–my–ability–to–perform–my–primary–function.
Job–o–Tron: I–will–miss–my–quota–and–be–downsized. They–may–even–replace–me–with–a–non-golem.
Job–o–Tron: Tertiary–objective–activated. Step–one: Acquire–object–designated–”hobo–bindle."
Job–o–Tron: Step–two: Proceed–with–new–career–as–a–wastrel–and/or–vagrant. "

The three laws of robotics:

1.A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2.A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.
3.A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Even though the first and second may not apply altogether to Golems made by the Consortium, there’s a possibility that having failed to fulfill its original intent, left the poor Tron with nothing else to aim for but fulfill the “Tertiary objective” at the best of its abilities, scrounging money for survival and seeking employment.

If that’s the case, then all of its current endeavors come from objeying a clear, pre-built subroutine alongside with logical assessments on the situations and actions dictated more-or-less by said cold logic (Such as offering to become Scarlet’s Henchgolem.) rather than true free will.

I’m not sure the three laws of robotics apply to GW2 golems. They’re not even used in most science fiction and certainly don’t apply to real life AI.

They were created by Isaac Asimov for his short stories and fairly importantly were created for the express purpose of sounding solid at first but being full of loop holes, exceptions, flaws and not applicable to all situations so that he could write lots of stories about how and why they didn’t work and what happened as a result.

However you’re right that we don’t know what Ho-Ho-Tron’s tertiary objective actually involves and it could be that everything he’s done since then is a result of that.

Although the article on his development also said losing his job made him go haywire, so it’s also possible he’s misinterpreting his programming. Like when the printer at work suddenly decides to print 3 pages of a document on the same side of one piece of paper, one of them upside down when all I asked for was a normal print out.

Sorta confirmed sentience here

https://forum-en.gw2archive.eu/forum/game/lore/Is-Hero-Tron-technically-lionguard-now

He is still an awesome character though :p

Is he sentient? Or is he just an exceptionally complex but ultimately just a mindless machine?
Would you know the difference?
Would it matter?

Something… Something… Turing Test