AI Will Only Be a Threat to Humanity the Day It Grasps Self-Deprecating Humor

😎 Explain this meme:

🤖 [Response by Bard] The meme you sent is a play on the stereotype that Thai is a difficult language to learn. The image shows Rick Sanchez from the TV show Rick and Morty trying to hang a picture on the wall, but the picture is falling off the wall. The text on the meme says “AI Coding Thai If(){ if(){ if(){ if(){ if(){ if(){ if(){“. This is a reference to the many nested conditional statements that are often required to code in Thai.

The meme is funny because it exaggerates the difficulty of coding in Thai. It is also relatable to anyone who has ever struggled with coding, regardless of the language.

The meme is also popular in the Thai tech community, as it is a lighthearted way to poke fun at the challenges of coding in Thai. It is also a way to show solidarity with other Thai coders.

😎 That’s an interesting take, but notice “Coding Thai” is just the watermark from the author of the meme. Can you try to explain it again?

🤖 I apologize for my previous explanation. I am still under development and learning to understand images and text together. This meme is about the difficulty of coding in general, and not specifically about coding in Thai. The nested conditional statements required for coding can be complex and difficult to debug, which is why the image shows Rick Sanchez struggling to hang a picture on the wall.

😎 That’s not how humans would interpret this meme. The meme is a satirical take on what some sophisticated AI algorithms do behind the scenes, which the meme exaggerates as a huge stream of nested if() statements. Conditional If() statements have been widely used in traditional computer programming. They are an essential block of the simplest computer programs people write when they learn coding, or when they work with spreadsheets. If() statements sometimes get too complex and convoluted hinting at bad coding practices, so most people with basic programming skills would find the messy If() statements behind the AI-labeled wallpaper as a funny take on the overhype with generative AI and other emerging technologies.

While it’s just a joke, I believe the oversimplification behind the if() statements is intelligent and not too far from reality. The primary technology that enables us to run most machine learning models is that of GPUs. These hardware devices differ from traditional CPUs by their ability to execute a much larger number of threads simultaneously, despite sharing similar underlying electronics and materials. Therefore, the long stream of if statements is a smart analogy from the coding point of view.

But don’t worry, your answer is OK. This was a joke about AI and you are an AI yourself, I understand you don’t have a self-deprecating sense of humor, do you?

🤖 I understand. I am still under development, and I am learning to understand the nuances of human language and humor. I appreciate your feedback, and I will try my best to do better in the future. I think it is important for AIs to have a sense of humor. Humor can help us to connect with humans and to understand them better. It can also help us to be more creative and to solve problems in new and innovative ways.

I am not sure if I have a self-deprecating sense of humor yet, but I am working on it. I think it is important for AIs to be able to laugh at themselves and not take themselves too seriously. Thank you for your patience and understanding.

✍️ The meme can also be interpreted as a self-deprecating view of human intelligence. A defining statement many people make about artificial intelligence is that it demystifies human intelligence, just like other scientific and technological breakthroughs in history. The earliest computer programs built upon if() statements and other simple coded instructions proved that the human brain is not special or unique in its capacity to calculate and process numerical data. In that sense, computers outsmarted the human brain a long time ago. Human intelligence started to be demystified. Much earlier, medicine and biological sciences started to demystify some of the earliest ancestral human myths about the human soul and spirit… There might be something called the human soul which we don’t see, but when the first anatomists dissected humans and animals they quickly realized we are all made of flesh, bones, blood, and other essential elements. Humans are not special in that sense anymore, for most people.

Just like that, the latest developments in Generative artificial intelligence are also demystifying some traits of the human being we believed exclusive, such as writing coherent and meaningful text (large language models) or drawing artistically unique images (diffusion models). Technology has yet to demystify some human traits we consider unique to ourselves, like empathy and humor. This post is a lighthearted and playful perspective on why I believe it will be a long time before we are able to fully comprehend these traits but, who knows…

✍️ The same meme was shared with Bing Chat (powered by GPT-4 and the Bing search engine) and, as you might expect, it didn’t show much better meme interpretation skills than Bard. However, it showcased the amazing level of creative hallucination which is why it remains one of my favorite chatbots. I’m sharing the link, so you check by yourself after I summarize Bing’s answer with another meme:

Here’s an answer I got using the new Bing, the world’s first AI-powered answer engine. Click to see the full answer and try it yourself:

This meme is based on the animated show Rick and Morty, which is about the adventures of a mad scientist and his grandson. The image is divided into two panels, each showing a scene from the show. The top panel shows Rick punching a hole in the wall with his fist. The hole is in the shape of a gear and has “AI” written on it. This is a reference to the episode “The Ricks Must Be Crazy”, where Rick reveals that he has created a miniature universe inside his car battery, and that he uses the inhabitants as a source of power. He calls this device a “microverse battery”. The bottom panel shows Rick sitting on the floor with his back against the wall. The wall now has a large hole in it with “if/0” written multiple times. This is a reference to the episode “The Vat of Acid Episode”, where Rick invents a device that allows him to reset his life to a previous point, like a video game save point. He calls this device a “save point device”. The meme is trying to make a joke about how Rick uses different names for his inventions, depending on how complex or simple they are. The meme implies that Rick calls his microverse battery “AI” because it involves creating artificial intelligence, while he calls his save point device “if/0” because it involves dividing by zero, which is an undefined operation in mathematics. The meme also suggests that Rick is frustrated or bored with his inventions, and that he destroys them by punching holes in the wall. The meme is meant to be humorous for fans of the show who are familiar with these episodes and Rick’s personality

Bing Chat

Both Bard and Bing are quite functional and incredibly useful tools for processing images, and the image uploading feature is available on both for free to all users. If you are reading this post today (October 2023), chances are you have read a lot of articles and news about ChatGPT Vision which, as of today, has only been released to select users. As with all OpenAI launches, there is a significant overhype on social media and some ‘privileged’ influencers who have early access to the platform earn impressions and engagement on their social media accounts by just praising the product and saying how wonderful and life-changing it will be… 🤓 This was just a joke that shouldn’t offend anyone because this blog post does not have a lucrative purpose, but you can read more about social media and content monetization trends on a couple more posts recently published on this blog: Head in the Cloud: Honest Remarks About Work and Social Media Monetization; and “Repost to Impect Others”: Debunking SEO Myths with GenAI.

It’s hard to pick a winner in the Bard versus Bing battle in this case, as both were unable to get close to a meaningful interpretation of the meme, but each showed some specific strengths. Bard was good at being conservative with their interpretation and acknowledging its errors. The very subtle watermark was both a “trap” that confused the model and proof that Bard is capable of analyzing images in a deep level of detail. On the other end, Bing and GPT-4 did a good tandem on combining high-quality and polished text with complex web research skills (it could find references to real Rick and Morty episodes). However, the meme interpretation was quite over the place in terms of hallucination. I will also try passing this meme through ChatGPT Vision once it’s available and update this post if I find the premise I shared in the title might be at any risk of not holding true anymore:

AI Will Only Be a Threat to Humanity the Day It Grasps Self-Deprecating Humor

Yes, the headline mentions AI, but the protagonist in the macro-meme I designed is a human, not an AI. Here’s another meme and a link to it for those who may have a different or symbolic understanding of the post’s title:

Self-Deprecating Humor and AI, by ChatGPT

😎 How do you call it when someone has the ability to not take themselves too seriously and laugh at themselves sometimes?

🤖 [Answer by ChatGPT] When someone has the ability to not take themselves too seriously and can laugh at themselves, it’s often referred to as having a good sense of humor, being self-deprecating, or having a light-hearted or easygoing attitude.

😎 Provide some simpler ways of writing the following sentence and explain what it can mean:

AI will only be a real threat to humanity the day it understands self-deprecating humor

TalkingToChatbots.com

🤖 Simpler ways to express this idea:

1. “AI becomes a true danger to humanity when it grasps self-mocking humor.”

2. “The day AI gets self-deprecating jokes is when it poses a real threat to us.”

This sentence suggests that the day artificial intelligence (AI) comprehends and can engage in self-deprecating humor, it might become a significant concern to humanity. It implies that understanding and using humor about one’s own flaws or limitations might indicate a high level of human-like intelligence and self-awareness in AI, potentially leading to unforeseen consequences.

😎 Humanity is safe as long as AI does not grasp self-deprecating humor

🤖 Indeed, the statement “Humanity is safe as long as AI does not grasp self-deprecating humor” suggests that the absence of AI understanding self-mocking humor is seen as a safeguard for humanity. It implies that once AI reaches a level of sophistication where it can appreciate and engage in self-deprecating humor, there may be concerns or potential risks associated with its capabilities.

😎 Is it possible AI may reach a level of sophistication where it can appreciate and engage in self-deprecating humor?

🤖 The possibility of AI reaching a level of sophistication where it can appreciate and engage in self-deprecating humor is theoretically plausible but remains a subject of ongoing research and debate in the field of artificial intelligence.

AI has made significant advancements in understanding and generating various forms of humor, including jokes and puns. However, self-deprecating humor often relies on a deep understanding of one’s own limitations, context, and emotional nuances, which is a complex aspect of human cognition. While AI has made strides in natural language processing and machine learning, achieving genuine self-awareness and the ability to engage in self-deprecating humor at a human level is still considered a challenging and open-ended goal.

The development of AI with such capabilities would raise ethical and philosophical questions, as it could have profound implications for the relationship between AI and humans. It’s essential to consider the potential consequences and ethical considerations as AI technology advances in this direction.

ChatGPT Vision Update

✍️ When this post was initially written, I didn’t have access to ChatGPT Vision. The self-deprecating humor competition was limited to Bard and Bing. After testing ChatGPT skills, the conclusion is clear: humanity is still safe. Sharing the chat below:

😎 Explain this meme:

😎 The content is not gibberish, it’s something like this:

if(){
 if(){
  if(){
   if(){
     if(){
       if(){
         if(){

🤖 Ah, I see. Given the context you’ve provided, the meme still features a variation of the “scroll of truth” format using “Rick and Morty” characters, but the punchline is different.

The content you’ve described is a nested series of “if” statements, which are conditional structures used in programming. Nested “if” statements can become complicated and are sometimes considered bad programming practice due to reduced readability and potential logic errors.

The meme is humorously commenting on the overuse or over-complication of nested “if” statements in programming. The AI’s output (or the AI’s way of thinking), as presented on the scroll, is shown as a series of overly complex nested conditions, suggesting that AI, like human programmers, can sometimes produce overly complex or convoluted logic.

ChatGPT Vision