AI Ethics Fervently Frowning About Those AI Sentience Jokes That Are No Laughing Matter

They say that laughter is the best medicine.

Except perhaps when the humor meant to generate laughter is unwittingly construed as bone-dried seriousness. In that case, instead of therapeutic laughter, there can be confusion, consternation, and potential sorrow.

An especially outsized exception of not being curative occurs when the misinterpreted humor is outspokenly employed by someone in authority or that carries a lot of societal weight. That is an important point that I am going to hammer away on in this discussion.

The altogether misunderstood soothsayer inadvertently or intentionally says something that can seemingly rock the world and is believed to be outright sincere and significant. The person uttering the humorous joke or anecdote might be entirely unaware of how serious their utterance is being taken. They sit in their own blissful humor-laden bubble. Everybody surely will obviously discern that the expressed aspect was meant in jest, so the person cheekily assumes. Nobody could see things any other way. People will certainly be rolling on the floor in completely impulsive fits of laughter.

Maybe not.

Before I take you through the challenges associated with humor that gets misunderstood as seriousness, I’ll spill the beans as to why I am taking you through that bumpy dusty trail. Turns out that AI Ethics is worried about and frowning on those that are perceived to be knowledgeable about AI that decide to make jokes about AI sentience. As I will momentarily explain in detail, such jokes are confounding and raise angst, and are said to be detrimental to the public at large. For my ongoing coverage of AI Ethics and Ethical AI, see the link here and the link here, just to name a few.

First, let’s get back to the general tale of woe about basic humor that gets miscast.

Current headlines vividly illustrate the type of disconnect that can arise.

Consider the recent assertion by Elon Musk that he was going to buy the Manchester United soccer team (spoiler alert, the remark was meant as a joke), in which he tweeted: “Also, I'm buying Manchester United ur welcome”.

His seemingly just-saying tag-along tweet of about seven words in length went around the world nearly instantaneously and raised quite a ruckus. Why was he truly intending to buy the soccer team? Would he be pouring tons of money into the team? Might he opt to start gobbling up sports teams across all manner of sports franchises? Is this a new business angle for him or was it a bit of idiosyncratic whimsy?

The Internet went into one of its usual digital hand-wringing convulsions and a newsy storm erupted.

About two hours later, he tweeted that his remark was in jest.

Also, he stated that he is not buying any sports teams. Plus, in order to apparently soften the blow to the Manchester United team, since they presumably had moments of a rush of grand excitement amid contemplating such a purchase by the wealthy provocateur, there was an added comment that if he ever did buy a team it would be them. I guess this is some solace for the players. They will forever know in the back of their minds that he might someday buy them, assuming that he opts to buy a sports team and assuming that he keeps his word.

Probably best to not hold one's breath on that house of cards.

In any case, you can almost shrug off this particular example of humor that went bust.

Musk is widely accustomed to making humorous remarks that go haywire due to people believing the comment to be serious and sincere. Part of the problem is that it isn’t easy to know when he is making a brazen statement that is meant with abundant sincerity versus the ones that are idly intended as mirth.

You cannot know for sure.

The added twist is that sometimes he says things that he suggests are jokes, but they turn out to not be. His style is pretty baffling. Serious remarks that turn out to be jokes. Jokes that turn out to be serious. At times, he makes remarks that appear to live in both worlds at the same time, existing as both jokes and seriousness simultaneously (maybe due to his familiarity with Schrödinger’s cat and quantum mechanics, see my analysis at the link here).

Would we care about his remarks if he didn’t perchance happen to be a mega multi-billionaire and was already known for making seemingly outstretched purchases?

I doubt we would.

The gist is that because he is in a position of authority or otherwise has societal hefty weight, a humorous anecdote can have tremendous repercussions if thought to be serious.

Keep your eye on the ball, namely that the stature of the jokester is remarkably crucial.

Do a quick compare and contrast on this notion. If a standup comedian that hosts a late-night talk show makes allegedly funny comments, we are likely to downplay that the remarks are meant to be serious. We know they tell jokes. It is their normalcy. Even this though has its limits. A comedian that appears to be serious and no longer seems in a wink-wink comedic mode can get themselves into pretty hot water about jokes that go flat.

When a person that is known for both seriousness and comedic fare tells jokes, things can get really dicey. Particularly when the person is perceived as being able to enact or bring to fruition whatever they are joking about.

You might recall that when Musk first indicated he might buy Twitter, there was all manner of confusion. Could he afford it? Yes, there is little doubt that he could. Would he want to do so? Yes, he had relentlessly complained about how Twitter was being run, implying that he could do better. And so on. There were few telltale clues available to figure out whether the matter was a joking one or a serious one (or both?).

I guess you could say we still don’t know.

When a person in authority has a longtime history of being strictly serious, the odds are that anything they utter will be presumed as being straightforwardly sincere. The thing is, it doesn’t even enter into our minds that their remarks are intended potentially as a joke. Indeed, if they try to tell a joke, their only hope of getting laughter would be to loudly and boldly proclaim beforehand that they are about to tell a joke. Without the heads-up declaration, anything they say is going to be bereft of laughter due to the ironclad belief that they always are straight as an arrow in their remarks.

Okay, we have these rules of thumb:

• Jokes can sometimes be misconstrued as seriousness
• Misconstruing can lead to adverse consequences
• The consequences are more likely pronounced by the stature of the joke teller
• Style of the joke teller matters a lot too
• A high-stature known-jokester has some latitude available
• A high-stature always-serious person has a minimal latitude
• A high-stature semi-jokester sits in a muddied middle ground
• Low-stature emitted misconstruing remarks are usually not given much weight

I’d like to share one additional recent example of a bad choice in joke telling.

To start with, you might vaguely be aware that the famous James Webb Space Telescope (JWST) made its global debut in the sense that images from this amazing marvel of science were finally publicly released in July 2022.

The Webb Telescope is considered to be the successor to the Hubble space telescope. What you might not readily know is that this was a moment long in coming. The initial design of the JWST began in the mid-1990s. At times, the effort had a somewhat tortuous path to reality. It also cost many billions of dollars to build. Furthermore, there was an enormous risk that the contraption might not make it into space, blow up at launch, or might not beam back usable images. All manner of things could have gone wrong.

This was serious business, as they say.

Some doubted that the Webb Telescope would ever function. Many were dismayed that so much money went to this particular scientific endeavor. Perhaps the monies ought to be spread around to other science projects, some insisted. Controversy dogged the project throughout its over 25 years of coming into being.

I trust that you grasp where I am going.

The Webb Telescope is a serious matter. That being said, trying to tell jokes about the Webb Telescope had become commonplace for those that felt the project was a big mistake. But those jokes were downplayed because they were usually lobbed by those known to be critical of the endeavor. You knew that when they joked, it was a joke, though expressed with a serious undercurrent of not wanting the Webb Telescope to proceed or that some other alternative was preferred instead. Those jokes have now pretty much fallen by the wayside since the spacecraft is up and running.

The last thing you would ever imagine is that someone that might have been seen as an avid supporter of the Webb Telescope would lend themselves to being a jokester about this latest and greatest in space exploration. That is your setup for the recent news item of relevance here.

According to news reports, a prominent French scientist released a photo that he stated was taken by the Webb Telescope and which purportedly provided a stunning new image of Proxima Centauri (considered the closest star to the Sun). The image seemingly showcased an astonishing level of eye-popping detail heretofore never seen by humanity.

You would certainly assume without hesitation that the image was indeed produced by the Webb Telescope. A notable element of that assumption would include the fact that the French scientist was well-known for his work in this field. He also serves as the director of research for the French Atomic Energy Commission. Simply stated, he is someone with relatively high stature or authority when it comes to this topic.

A few days after having posted the image, he surprised the world by indicating that it was all a gag.

Turns out that the photo was actually of sausage chorizo that was staged on a black background. It wasn’t at all an image from the Webb Telescope. He had sneakily contrived an image that would reasonably appear to be from the far reaches. Of course, his having claimed that it was from the Webb Telescope was the kicker since his stature would carry a sufficient weight that nearly any “reasonably” apt-looking image could be said to be from outer space.

Why did he do it?

He indicated that it was both a form of amusement and that he was trying to convey an important message about being wary of blindly accepting the word of those in authority.

An outcry arose about this tomfoolery. Given that society is already confronting a plethora of misinformation and disinformation, his choosing to make a point about a related facet seemed ill-timed. The other side of that argument, some expressed, was that it was well-timed since it serves as a timely reminder to be wary of what you read and hear.

Do you think he was right to pull off this trickery?

Meanwhile, some have pointed out that his released staged image will live on.

We all know that items posted on the Internet take on their own lives, as it were. The odds are that the image will continue to be shared across the globe. People that see the image won’t necessarily know that it was staged. The original indication that the image is from the Webb Telescope will undoubtedly remain intact. You might anticipate that the photo will ultimately appear in textbooks and be posted on numerous websites as a close-up of sorts of the Proxima Centauri.

In that sense of things, his attempt to teach us all a lesson is not cost-free. There is a strong chance that the image will be spread wide and far, dispersing the lie and taking in many that would have no idea it is a lie. You might quibble with the contention that it was said to “merely” be a joke, but in the end, it was a lie. A lie that was spawned as part of a joke. The joke aspect might or might not be valuable. Nonetheless, the lie (or joke) is going to have reverberating repercussions, particularly due to the globally interconnected world and the vastness of the Internet.

To revisit the earlier stated rules of thumb, see how some of them are aptly supported by this Webb Telescope imagery fakery story:

• Jokes can sometimes be misconstrued as seriousness – a scientist releases claimed image of Proxima Centauri
• Misconstruing can lead to adverse consequences — public and researchers assume it to be true
• The consequences are more likely pronounced by the stature of the joke teller — prominent scientist carried out the joke or fakery
• Style of the joke teller matters a lot too — generally known for an upstanding scientific reputation
• A high-stature known-jokester has some latitude available — he was not known as a jokester per se
• A high-stature always-serious joke teller has a minimal latitude — this seems to be this instance

If a non-scientist tried to undertake this “joke” they probably would have a hard time gaining traction since no one would especially believe or assume that the non-scientist knew what they were doing or showcasing. The reputational cloak of the prominent scientist made this joke feasible.

We should also take a moment and ponder the motives underlying the joking.

The faked Webb Telescope image was part of a shall we say elaborated joke, given that the image took some effort to create as fakery. There is no question that this was a deliberate effort. When Musk made his tweeted comment about buying the soccer team, this does not seem to be quite on the same level of preparedness, nor did he afterward attempt to justify the joke as being a means of providing a lesson to the world about being mindful of what you read or hear.

All in all, I think most of us would likely take a dimmer view of an elaborately concocted joke that supposedly was to provide a life lesson versus a joke that was one-off and ad hoc. Too, we need to consider the result of the joke. Was the Webb Telescope life lesson worthy of the joke that pulled the wool over our eyes?

Playing jokes can be a complicated gambit and not necessarily excused simply because the jokester later contends the joke had an important purpose. On top of this, there is always the chance that a claim of having made the joke for a life lesson could be an afterthought rather than the core essence. We cannot sometimes know for sure that the joke was set up as a joke rather than trying to, later on, excuse the matter by a hand waving that it was all meant to be a joke. That serves as a common copout and might be a wild attempt at a saving grace.

Now that we’ve gotten the beguiling riddle of jokes and joke-telling onto the table, we can move into the realm of jokes about AI.

Before I delve into the serious topic concerning AI-related jokes and joke-telling, it might be useful to clarify what I mean when referring to AI. There is a great deal of confusion as to what AI connotes. I would also like to introduce the precepts of AI Ethics to you, which will be integral to this discussion.

Stating the Record About AI

First, let’s make sure we are on the same page about the nature of today’s AI.

There isn’t any AI today that is sentient.

We don’t have this.

We don’t know if sentient AI will be possible. Nobody can aptly predict whether we will attain sentient AI, nor whether sentient AI will somehow miraculously spontaneously arise in a form of computational cognitive supernova (usually referred to as The Singularity, see my coverage at the link here).

Realize that today’s AI is not able to “think” in any fashion on par with human thinking. When you interact with Alexa or Siri, the conversational capacities might seem akin to human capacities, but the reality is that it is computational and lacks human cognition. The latest era of AI has made extensive use of Machine Learning (ML) and Deep Learning (DL), which leverage computational pattern matching. This has led to AI systems that have the appearance of human-like proclivities. Meanwhile, there isn’t any AI today that has a semblance of common sense and nor has any of the cognitive wonderment of robust human thinking.

Part of the issue is our tendency to anthropomorphize computers and especially AI. When a computer system or AI seems to act in ways that we associate with human behavior, there is a nearly overwhelming urge to ascribe human qualities to the system. It is a common mental trap that can grab hold of even the most intransigent skeptic about the chances of reaching sentience. For my detailed analysis on such matters, see the link here.

To some degree, that is why AI Ethics and Ethical AI is such a crucial topic.

The precepts of AI Ethics get us to remain vigilant. AI technologists can at times become preoccupied with technology, particularly the optimization of high-tech. They aren’t necessarily considering the larger societal ramifications. Having an AI Ethics mindset and doing so integrally to AI development and fielding is vital for producing appropriate AI, including the assessment of how AI Ethics gets adopted by firms.

Besides employing AI Ethics precepts in general, there is a corresponding question of whether we should have laws to govern various uses of AI. New laws are being bandied around at the federal, state, and local levels that concern the range and nature of how AI should be devised. The effort to draft and enact such laws is a gradual one. AI Ethics serves as a considered stopgap, at the very least, and will almost certainly to some degree be directly incorporated into those new laws.

Be aware that some adamantly argue that we do not need new laws that cover AI and that our existing laws are sufficient. In fact, they forewarn that if we do enact some of these AI laws, we will be killing the golden goose by clamping down on advances in AI that proffer immense societal advantages. See for example my coverage at the link here.

In prior columns, I’ve covered the various national and international efforts to craft and enact laws regulating AI, see the link here and the link here, for example. I have also covered the various AI Ethics principles and guidelines that various nations have identified and adopted, including for example the United Nations effort such as the UNESCO set of AI Ethics that nearly 200 countries adopted, see the link here.

Here's a helpful keystone list of Ethical AI criteria or characteristics regarding AI systems that I’ve previously closely explored:

• Transparency
• Justice & Fairness
• Non-Maleficence
• Responsibility
• Privacy
• Beneficence
• Freedom & Autonomy
• Trust
• Sustainability
• Dignity
• Solidarity

Those AI Ethics principles are earnestly supposed to be utilized by AI developers, along with those that manage AI development efforts, and even those that ultimately field and perform upkeep on AI systems.

All stakeholders throughout the entire AI life cycle of development and usage are considered within the scope of abiding by the being-established norms of Ethical AI. This is an important highlight since the usual assumption is that “only coders” or those that program the AI is subject to adhering to the AI Ethics notions. As prior emphasized herein, it takes a village to devise and field AI, and for which the entire village has to be versed in and abide by AI Ethics precepts.

I believe that I’ve now set the stage adequately to examine more the sobering topic of AI-related jokes and joke telling.

AI Sentience And Jokes Thereupon

You can readily see or hear AI-related jokes just about everywhere.

The popular Dilbert cartoons seem to be a particular favorite at making AI jokes. A typical portrayal consists of making some pointed commentary regarding sentient AI. For example, a robot is walking around and seems to be sentient. One of the human characters interacts with the sentient AI. Invariably, the sentient AI is made to look stupid, or the human characters are made to look stupid. It is all good, clean, and generally purposefully lessons-learned comedic fare.

Everyone that happens to see such a cartoon knows that it is a cartoon. The fact that a joke is being conveyed is an obvious given. If a joke isn’t conveyed, the chances are that the person reading the strip would be disappointed and upset.

Let’s move outside of the comedic realm.

Suppose that a prominent AI researcher or scientist makes a pronouncement about AI sentience. Assume further that the statement is made in seemingly erstwhile seriousness. We assuredly would take this with notable importance. We would also want to check the track record of the person to see whether they are known for wisecracking or not.

There is a case in hand that illustrates this.

Recent news headlines blared that a Google engineer had encountered sentient AI while using a system known as LaMDA (Language Model for Dialogue Applications). According to the developer, he genuinely believed that the program had attained sentience. Indeed, the programmer asked the app whether it was sentient, and it fully and unequivocally said that it indubitably was.

Please be aware that the developer was completely mistaken and the program had absolutely inarguably not reached sentience, see my detailed coverage at the link here.

In this case, the fact that the engineer was with Google provided the stature-inducing aura. The assumption by society at large was that anyone working at Google would of necessity be of the highest caliber and that the assertion is presumably sanctioned by Google (it wasn’t).

Was the engineer playing a joke on us?

Apparently not.

The engineer continues to insist that the program is sentient.

One supposes that at some later point in time the engineer could surprise us by claiming that it was an elaborate joke, but this doesn’t seem to be in the cards. I mention this possibility since there is always the societal escape clause that can be used, namely later on recasting something such that you say it was a joke all along (I briefly noted this earlier). As noted, this sometimes works. In other instances, this won’t work. Even if it works, there can be a lot of lambasting that a joke was played, especially if it was of notable untoward consequences and if it was kept going for an extended period of time.

Though the news item doesn’t precisely fit our focus on AI sentience jokes since it wasn’t meant as a joke (seemingly), it nonetheless fits the essence herein that remarks about AI sentience are particularly given attention if made by someone in a position of stature and if they do so with a seriousness involved.

Not all remarks about AI sentience will garner outsized attention.

Let’s explore the nature of AI sentience comments that are generally likely to arouse interest if they are made by someone presumed of relevant stature and if made seriously by someone that is relatively assumed to be serious (i.e., not a comedian or prone to comedic outbursts).

We will categorize AI sentience assertions into three convenient buckets:

• Time-based unequivocal assertions about AI sentience
• Outcome-based unequivocal assertions about AI sentience
• Characteristics unequivocal assertions about AI sentience

Notice that I refer to the assertions as being of an unequivocal nature.

I mention this keystone criterion since a waffling or seemingly loose assertion is going to be less impactful than one that seems to be entirely without any qualification. If I tell you that the sky is always blue, this is a rather striking assertion and states without reservation that the sky is forever blue. If I say instead that the sky is sometimes blue, this is not quite as dramatic of a claim. An out is provided in that if the sky isn’t blue, that wobbly assertion still works.

We’ll see in a moment how this applies in the AI sentience proclamations.

Let’s next consider some prime examples of AI sentience assertions as classified per my three mentioned categories:

Time-Based Unequivocal Assertions

• AI sentience is here
• AI sentience is nearing
• AI sentience will occur on this date
• Other

Outcome-Based Unequivocal Assertions

• AI sentience will destroy humankind
• AI sentience will enslave humanity
• AI sentience will save humankind
• Other

Characteristics Unequivocal Assertions

• AI sentience will be super-intelligent
• AI sentience will be omniscient
• AI sentience will be emotionless
• Other

Perhaps the most grandiose of the AI sentience assertions is listed in the time-based unequivocal assertions and entails claiming that AI sentience is here.

That is the assertion that was made by the Google engineer. And it suitably got a lot of global attention since it is quite the assertion to be made. We are all waiting with bated breath for the day that AI sentience is attained. You could almost put this on the same footing as waiting for the day that an alien from outer space or a UFO is asserted to be truly here, see my related coverage at the link here.

Returning to the emphasis about these assertions being unequivocal, suppose an assertion is made by some prominent AI developer that isn’t so clear-cut. Perhaps the assertion is that they might have devised AI sentience, but they aren’t yet sure of it. Would that be quite headline-worthy? Probably not. It could still abundantly grab attention, but the assertion is not unequivocal and therefore allows for wiggle room. People would be aware upfront that the assertion might or might not be valid.

Continuing further into the time-based AI sentience assertions, another popular one is that AI sentience is nearly here.

Admittedly, this is not entirely an unequivocal assertion if there isn’t some definitive time frame stated. A person uttering this claim could have lots of outs. Does nearly here mean by tomorrow? Or does it mean within a year from now? Or maybe it means in the next thousand years. You can see how the nearness is ambiguous unless something else bounds the wording.

Despite the nearness looseness, this still can be an AI sentience assertion that generates quite a kerfuffle. If the person making the claim is of sufficient stature, and if they are generally known for being serious, the world at large is likely to take at face value that near means relatively nearby. We might not know whether this is within the next day or weeks, but most would ordinarily imagine that near means within the next few years.

Go ahead and take a look at all of the other AI sentience assertions that I’ve listed in each of the three categories and contemplate them mindfully. I’m betting that you’ve seen or heard them before. Plus there are others that can be easily added to the lists.

The focus here is whether those assertions are made for joking purposes.

We have the Google engineer instance that was apparently not made for joke-telling reasons. But the matter still remains about the arising ethical questions of whether making such a proclamation is appropriate. We can draw on the precepts of AI Ethics to aid such an assessment. Even assuming that the engineer was fully and unequivocally serious and sincere, should such declarations be made to the public at large?

You could argue that yes, of course, such a proclamation needs to be made to the world. If someone earnestly believed with all their heart and soul that AI sentience had arrived and they had “proof” directly in their hands, it would almost seem improper to not tell the world about it. We would all want to have a heads-up. We would all want to figure out what we should do. We would want to ascertain what the AI sentient “being” is going to do, such as whether it is going to aid humanity or try to destroy all of us.

Wait for a second, the retort comes, this can be akin to crying wolf. Until you have double and triple-checked things, going around and proclaiming that AI sentience has been reached is a pretty big deal. Especially when it comes from someone of a presumed stature that appears to be serious about the claimed assertion.

Let’s now consider the case of a purposeful joking assertion about AI sentience.

Consider the following scenario.

Some prominent AI researcher or scientist announces that AI sentience has been achieved. They do this with a straight face. No funny business seems to be afoot. Later on, they reveal that it was all a joke. It was done partially for fun and partially to provide a lesson. The lesson was that the public need to be careful in believing what they read or hear.

What do you think of that joke?

What do you think of that joke teller?

On a crying wolf basis, the public is going to gradually become desensitized to talk of AI sentience if these kinds of bogus claims are increasingly made. You could argue that this is bad for society since people are going to get confused about whether AI sentience has or has not been attained. They might simply remember that there had been a lot of talking about it, and not connect the fact that it was a joke.

But that’s just with regard to the assertion that AI sentience is here. There are also unequivocal assertions that AI sentience will arise in the years 2030, 2040, 2050, and so on. There are assertions that sentience AI will unequivocally wipe out all of humankind. Etc.

If those are based on joke telling, rather than on serious and insightful discourse, AI Ethicists are justly worried about the watering down of societal attention to what otherwise should be vital topics to examine. Anyone that seriously tries to delve into such a topic will suddenly find themselves mired in the joke-telling versions. People will assume that this is just another prank or prankster trying to be funny.

Should we put the kibosh on all high-stature serious-looking jokes about AI sentience?

You might be surprised to know that some vehemently argue in favor of telling these AI sentience jokes.

The rationale is that AI sentience is so far from reality that we should not be giving it serious attention now. The more the merrier to get the world tuned out of the crazy contentions. Showcase the clowns for being clowns. All told, turning any discourse about AI sentience into a joking matter will put aside the otherwise futuristic and purely theoretical ambiguous and confounding debates about sentient AI.

Furthermore, the logic goes, that we should be focusing precious world attention on the here and now of contemporary non-sentient clunky AI. We have plenty to deal with. AI systems that are jampacked with undue biases and discriminatory embodiments are much more worthy of today’s attention. Stop wasting your energies on sci-fi hypotheticals and let’s get down to earth and deal with the emerging morass of AI For Bad, they exert.

Conclusion

A sentient AI and a human walk into a bar.

Whoa, that sounds like the start to a great joke.

If the remark is made by a comedian at a nightclub while up on stage, perhaps the joke is a dandy and will garner a huge laugh. But when those of AI stature make such a joke and do so in a manner that seems fully serious, we might raise our eyebrows and be somewhat dismayed. If the person is not known for their witticism, this raises worries even more so.

AI Ethics is keeping an eye on how far the AI sentience joking is going to go.

Some might think this seems like party poopers that want to place a wet blanket over some good-natured fun. Does wanting to moderate the telling of jokes about AI sentience become a grinch-like spoilsport? Or is this AI Ethics propensity serving as the thank goodness designated driver and wanting to make sure that the public is not falsely misled or endangered?

This is more than a lighthearted laughing matter.

One final note. Sentient AI might eventually look back at these times and relish all those AI sentience jokes, or might instead decide that the smarmy jokes were in bad taste and seek to discipline the joke tellers.

Makes you think twice about telling those jokes, doesn’t it?