Usually, by this number of writings on a topic, interest has started to wane. However, I don't see that happening yet on the topic of Artificial Intelligence. Besides, I think it's important I (and my readers) have a well-informed idea of what's involved here (if for no other reason than our own preservation/safety in a dangerous world, including ones where opportunists are jumping on the bandwagon to make a quick buck). To that end, I'm going to keep going with it.

The aim of the technology involved with Deep Learning is to imitate how the human brain learns. This is only possible because of the availability of fast and powerful computers, better algorithms, big data gathered as a consequence of the massive digitisation of our society, massive investments and unscrupulous practices of Big Tech companies engaged in surveillance capitalism (Amazon, Facebook, Google, Microsoft, Verizon, Yahoo! and others) and psychological manipulation for their own business interests.

Although people might claim that the AI winter is over, it's likely only temporarily so and another one will surely ensue at some point. That's because proponents of AI (primarily those intending to profit from it) are falling into the same old trap of fueling the hype machine and would do well to keep the rhetoric and promises to a minimum. While AI certainly can do amazing things, those things are still quite mundane in comparison to what humans can do. In this post, I'll be considering the current (and likely future) uses of AI. Perhaps we have an over-reliance on AI that isn't justified and would do well to reclaim some tasks from it. (If Elon Musk and Stephen Hawking are wary of AI, then I think there may still be some merit in me still being in the same position, despite what assurances proponents may offer. From what I can tell, AI being an existential threat to humanity on a massive scale is unlikely to be a real thing for at least fifty years, by which time I'll hopefully be dead or so thoroughly sick of life as to be welcoming gunshots by computer. Thank you, RZNR 10 000! As it stands, it's just the ones in antisocial media sites making individual people addicted to feeling terrible about themselves and suicidal. Nothing to worry about there, right? You can just log off and not go back, after all!)

The Uses of AI

One of the problems with just how pervasive AI usage is is that it's so thoroughly and well integrated into existing technology that it's difficult (if not impossible) to tell that it's already there. I would never have guessed that it was manipulating people with accounts on Instaharm, Faceborg and Scroogle if I hadn't read up on it. if you've got a smart device (phone, thermostat, TV, whiteboard or even home, AI is monitoring you and learning your habits in order to push you into being a consumer drone devoid of privacy.) Given that even most cars have computerised components, there's probably AI in your car. YOu can be sure that it's certainly prevalent in the workplace. The scary thing is that it's all hidden away from you, even (or perhaps especially) when it has a dramatic and/or profound impact on your life. (Most companies likely won't tell you that they use AI or for what purposes.) The number of ways in which AI is currently used is possibly in the millions and thus I cannot feasibly provide an exhaustive list. Here, then, is a small sample of the prominent ways in which AI is part of the technology of your life:

• Fraud Detection: If you've ever had a call or message from a bank or credit card company asking to confirm that you did actually make a particular purchase, then you can be sure that there wasn't someone at the company poring over your transactions; an AI picked up an irregularity in your spending pattern and alerted somebody. Maybe you might think that's a good thing and an acceptable use of AI, or maybe it's time to find a DEX and merchants that accept crypto payments. As things currently stand, I'm looking to get rid of my Solana (SOL) and have nothing further to do with the project because it uses Chat GPT ("glorified autocorrect", as Adam Conover calls it) for customer support interactions.
• Automation: The use of AI for handling unexpected changes and/or events in an automated process is very common. An item or object where it's not supposed to be can cause a process to halt (and maybe catch fire, depending on the nature of the process). AI can handle this erroneous occurrence and either restart the process or keep it running smoothly.
• Customer Service: When you call the Customer Service Hotline, it's very unlikely that you'll deal with a human agent most of the time. The theory is that an automated agent is sufficiently capable of handling the majority of queries and questions by following scripts. However, that's often not the case. Recently, I've experienced this more often than I used to and it frustrates the hell out of me. I blame the current hype around AI and companies not investigating the limitations of the technology in their rush to put it in place and retrench their Support staff. Having also worked in various Support Staff roles, I've got more than one reason to be opposed to such foolishness.
• Safety Systems: Many of the safety systems in machinery and vehicles rely on AI to assume control in a crisis. Automatic braking systems are a case in point. I find that somewhat ironic.
• Machine Efficiency: AI can be used in control systems, in order to obtain maximum efficiency from a particular machine. AI can place limits on speed and power consumption in order to meet certain efficiency goals.

Not Fueling the AI Hype Machine

"AI is BS!

Artificial Intelligence is a real field of computer science that's been studied for decades, and in recent years, it's made major strides, but I'm not talking about that kind of AI. I'm talking about the marketing term AI that tech companies are using to hype up their barely functional products, all so they can jack up their stock price.

See, tech companies are powered by hype. It's not enough to be profitable. No, in tech, you have to be able to convince investors that you have cutting-edge disruptive technology that will let you dominate an entire industry."
 — Adam Conover; Comedian

Having worked as a software and Web developer for at least two multi-national corporations, I can assert that his assessment of the industry has validity.

As is usual with The Next Big Thing™ in tech, there is a lot more hype around AI than there are objective use cases AI is actually capable of fulfilling. Popular media (particularly films and television) tend to add fuel to the fire and perpetuate the myth that AI development has progressed further than it actually has. In reality, AI is far from the level of sophistication and intelligence portrayed in popular media. (It's easy to imagine and portray something when you're not constrained by practical limitations on its implementation. Otherwise, we'd all be aboard the US Enterprise and traveling between galaxies at light speed. It's just not going to happen in this lifetime and probably not the one after that, no matter how many psychoactive drugs anyone takes.)

"I'm sorry, Dave; I can't do that." — HAL 9000; 2001: A Space Odyssey; Stanley Kubrick and Arthur C. Clarke

Captain Kirk has taken too much LSD!

You might have heard of something known as The Singularity. What this is is an algorithm that encompasses all technological aspects of AI. (It's the one AI to rule them all, including us, the computer equivalent of the Universal Consciousness, but you can be sure some people will still connect through IRC). It's that concept (which we're unlikely to see even starting to take shape until 2045) which inspires the futuristic super-AI depicted in films and television.

For The Singularity to occur, computers have to be able to learn and think like humans do. Currently, they don't because they can't; they aren't proficient in the types of learning specified in a prior post. When it comes to machine/deep learning, there are a number of factions or special interest groups (SIGs), usually called "tribes", working in particular areas:

• Symbolists: This group is primarily focused on matters of logical, mathematical/scientific and philosophic reasoning, using inverse deduction to solve problems. (In contrast to the usual/typical model of starting with a premise and looking for the conclusions, inverse deduction starts with a set of premises and conclusions and works backward to fill in the gaps.)
• Connectionists/Connectivists: This group is primarily focused on neuroscience (neural connections) and re-engineering the human brain, using back-propagation (adjusting weights and biases as a neural net learns) in order to solve problems.
• Evolutionaries: This group focuses on evolutionary biology, using genetic programming (evolving best-of-breed algorithms in generations) to solve problems.
• Bayesians: No, these are not members of an ancient civilization (as opposed to Byzantines). This group is interested in statistics and probability theory. They develop expert systems and use probabilistic inference to solve problems.
• Analogisers: This group is primarily interested in the application of human psychology and makes use of kernel machines (pattern analysers) in order to solve problems.

The Singularity will incorporate learning capabilities and aspects from all of these groups and their technologies (which requires a high degree of collaboration between them). That is their ultimate goal. Scary as that is, it is entirely possible that the five tribes might not actually be able to create algorithms that accurately or reliably imitate human intelligence. That doesn't stop them trying, however. The fact that there's much we still don't know about exactly how or why we think the way we do.

While we poor meat bags don't need to worry about computers achieving The Singularity and taking over the world for a while yet, it's going to happen at some point in the (hopefully distant) future. With that in mind, it would be great if those developing the technology took seriously such concerns as we have and implemented guidelines and safeguards to control, restrict and even override the technology while there's still time to do so. Nobody seems to be doing that, because it takes time away from developing a product that can be rushed to market and raking in profits from people who don't know any better, can't tell the difference between actual gold and gold plating or paint.

"In short, whether you're an AI pessimist, optimist, or somewhere in between, it's a good time to be wary of what exactly is being marketed to you, and how groundbreaking it actually is — or if it's just a half-baked cash grab."
 — Maggie Harrison; Futurism.com; 2023/04/08

Having done my own preliminary tests on the various AI options available online for generating artistic/creative output (notably music and visual content), I'm quite disappointed by the limited capability of what I've found. Mostly, though, there are products available for astronomical subscription prices, without so much as even a once-off trial or details of what results one can realistically expect. To me, those are the obvious cash grabbers.

As always, keep your wits about you and DYOR. Don't blindly trust black boxes and corporations, verify that they actually deliver on their promises.

Thumbnail image: Meat bags in Fallout 4, copyright Bethesda