Introducing the Disambiguation Podcast

We are excited to announce the official launch of the Disambiguation podcast.

"Disambiguation is the process of removing confusion around terms that express more than one meaning and can lead to different interpretations of the same string of text." 

Host Michael Fauscette of Arion Research; a leading technology analyst, tech startup advisor, consultant, board member, and storyteller; and his guests "remove the confusion around" artificial intelligence (AI), generative AI and business automation by looking at the business solutions available today to improve business outcomes and gain competitive advantage. 

The show is available weekly (Friday) on all the major podcast networks including Apple Podcast, Google Podcast, Spotify Podcast, Amazon Music / Audible, Pandora and TuneIn.

Here’s the transcript for Episode 0:

“Michael: Hello and welcome to disambiguation, a podcast about removing some of the confusion and helping you understand more about AI and business automation. Now we're once a week we'll bring interesting guests on to talk about all sorts of different topics around artificial intelligence and the adoption of artificial intelligence and how different companies are building products to provide solutions and some of the application of that technology and business itself.

The purpose is to try to cut through some of the noise that we're hearing. Certainly, over the last year, or at least since last fall there's been a great deal of interest in AI for business, particularly generative AI and of course anytime you see that much activity and that much excitement, there's also a certain amount of dark side or downside, or even just noise and understanding that needs to help you figure out how you're going to actually use these technologies in your business and, make some difference there.

Now I just conducted a survey on AI adoption, and I was just going through the data and one of the things that I think is really interesting is to look at how companies are using this technology today. The survey was North America only with 402 respondents. The number one use case at 65%, which is a pretty strong, is chatbots.

So intelligent chatbots are certainly popping up in a lot of different places, replacing your more traditional chatbot. 64% said cybersecurity, the number two use case in the survey. 53% are using AI for virtual assistants, 47% enhanced data analytics and 44% anti-fraud or, financial fraud prevention.

You can see strong adoption and use already. And of course, that's just growing. In that same survey, 87% of the respondents said that they would increase their spending on AI over the next 12 months. So there's definitely a need to understand and hopefully like I said, we'll be able to get some interesting guests in here.

We already have several things lined up. In upcoming episodes we're going to be talking about AI and security, we're going to be talking about privacy issues around AI, AI for marketing, prompt engineering,… you just can't go in and use these GPT engines without understanding how to write a prompt to get exactly what you want out of it…much like you had to learn how to search on Google when you started using search to get the results that you actually wanted. Sales and automation. Intelligent chat bots I mentioned already, and we'll dive into that a good bit. Data preparation, data quality, and data sources which is a certainly a hot topic. Not just your internal data, but how do you augment that data?

Using AI for cybersecurity is one of the hottest use cases, and of course, that really is a response to the fact that AI on the opposite side, on the hacker side or the darker side is also seeing a lot of adoption and use. And it makes the threat exposure much greater for businesses.

The best way to respond is to fight fire with fire, as they say. Generative AI use cases, we're all talking about. GPT engines and integrations and open and closed large language models. We'll be talking about a lot of that as well. And when we talk about artificial intelligence or AI, we're really talking about the simulation of human intelligence processes by machines, computers these processes are.

Things like learning the acquisition of information and rules for using that information reasoning, using the rules to reach approximate or definite conclusions and self-correction. We divide AI into three categories, narrow AI or sometimes called weak ai, general AI and artificial super intelligence.

What we're talking about today is in the category of narrow AI and we haven't yet been able to build a general AI system, and then certainly not as an artificial super intelligence, which would be even more capable than a human. So general AI or strong AI exhibits all of the capabilities of being human. That implies that it can perform intellectual tasks like humans can, to understand, learn, adapt, implement knowledge, et cetera. Like I said it's an exciting one, but it's definitely not out there yet. What we're talking about now is narrow AI and all the conversation today is around that.

For the near future anyway. It's hard to predict, but certainly there are plenty of people working on the move to general ai. We talk about AI, and I know a lot of people think that this is a fairly new development, but the truth is, if you look at the history of AI and I'm not going to take you through the whole history, although maybe we'll do a show on that at some point. But the field of AI really, emerged in the 1950s and sixties, which surprises a lot of people. And then it's very closely tied to things that we would've called analytics, for example. Alan Touring proposed the touring test in 1950 to determine the machine's ability to exhibit the intelligent behavior that's indistinguishable from that of a human.

And that's how we started moving forward with these. The Seventies and Eighties there were expert systems, knowledge engineering techniques to solve complex problems. Stanford cart became the first AI controlled autonomous vehicle in 73.

So some movement there. 1990s and 2000s, machine learning became the big topic. In 97, IBM's Deep Blue defeated world chess champion, Gary Kasparov, which really was the first time the public saw showcased the potential of AI and, in strategic decision-making kind of tasks.

And then, Natural language processing, which starts to help computers to understand and generate human language effectively and efficiently. In the 2010s, some developments there. We had IBM Watson on jeopardy, for example, and won Jeopardy showcasing the power there.

And the virtual assistants that we all started to use on a daily basis. Siri, Google Assistant, Amazon's Alexa, all of those are AI powered conversational agents that we started to use and be comfortable with. And much more commonly used and particularly, things that are built into your phone or some of the other smart home, your thermostat, et cetera. connect into these virtual assistants as well. So where are we going? That's a tough one. Certainly there's a lot of activity, there's a lot of experimentation, there's a lot of capabilities that are growing very quickly and with all technology, it tends, we tend to predict the future based on the rate of change that we have today. And the truth is, technology almost always advances exponentially. And this is certainly the case with the AI technologies. I suspect that you're going to see many leaps forward over the next few years. It's going to be a really exciting time.

And that's one of the reasons that we really wanted to launch this podcast now, so we could be inside of all the exciting things that are happening around these advances and rapid advances and all the ongoing research in areas like reinforcement learning, generative models, explainable AI, robotics, semi-autonomous vehicles, certainly trying to move towards autonomous vehicles, smart homes, healthcare, diagnostics so many business applications there. And then certainly we also want to take a look at some of the ethical considerations, concerning AI including things like issues of privacy, bias, job displacement, which I know is something where there is a lot of buzz lately, with the idea that perhaps the automation could eliminate certain job categories. When we publish the report from the survey, you could take a look at that. You'll see that there is a certain amount of displacement that's going to happen because you're going to automate over time, certain mundane tasks out of the way.

But that doesn't mean that you're eliminating jobs overall because there's certainly new skills that are desperately needed to implement and operate and continue to use these types of tools. That was the number one concern in the survey from the respondents, was this idea that it's very difficult to find the qualified talent that you need.

So there's going to be a big push to educate more people, get them involved in in AI research, AI application, use of AI. All those things are definitely going to, you'll see happening over the next few years. And then, of course, this general intelligence, the AGI, artificial general intelligence we were talking about is this idea of human level intelligence, which is still an aspiration at this point, but certainly will be moving along. Today for business use cases, you can think about two basic ideas. One is automation. So, what are the tasks and things, operational things that I’m doing in my business that I could automate. Right now we think of in the context of what we call mundane task or routine task, whatever word you want to use. But those are the things that are capable of being automated and that line over time's going to move up as more and more capabilities come out. More and more things that could be automated so you can free employees up to focus on much higher value kinds of activities. And then the other kind, and I mentioned assistance before, is the idea of using AI technologies as an assistant. Sometimes you'll hear people refer to enhanced data analytics, for example. The idea of having large data sets and having AI be able to parse through machine learning models, generative models, all those different types of technologies, parse through all this data and then narrow down your choices to help you make better decisions, whether it's a business decision to healthcare decisions, whatever.

In fact, that's one of the easiest use cases I think to understand when we talk about assistance is a doctor doing a diagnosis. We're not at a point where we want the AI to do the diagnosis, but you could feed that AI, all of the symptoms. And of course it trained on data from all over the world about different types of diseases and symptoms and that sort of thing. It could, pretty reliably, give the doctor a few choices or places to look to try to make the diagnosis and really assist that doctor in in. Consuming a mass data set that there's no way that a human would be able to work through in a reasonable amount of time and provide that diagnosis.

So that's the kind of thing that I think you're going to see more and more of assistance and automation. Going forward we'll take a look at some of those topics. We have some exciting topics coming up. One of the first shows will be about AI and marketing which is one of the areas we're seeing a really a lot of activity. We're going to look at customer experience in the context of AI. A few different ways to look at that. I have a few different guests around the idea of cybersecurity and AI. We're going to take a deep dive into several of those things. I did a show on my other show “In the Hot Seat”, not that long ago on generative AI in higher ed, and we're going to bring one of the guests from that show onto this show and have a conversation about what she's seeing in the application of gen AI in a higher ed environment. We've got some, we're going to look at some Salesforce activities at AI during Dreamforce.

So we have several other use cases to examine, virtual assistants, low-code / no-code platforms that are using the AI. We're going to look at that. And then of course, in the backend, the data quality, data sources, all those things. I'm excited to get this going. I don't want to ramble on too much today.

What I wanted to do was focus around just helping you understand what we're going to do on the show and where we're going to go forward with this. Look for the shows once a week in all your standard places if you'd like to see the videos, we will be doing videos for most of the shows, not necessarily all of them, but for most of the shows we'll do the videos and we'll be doing those shows for some of the onsite type shows, like Dreamforce, so I'm going to try to record a couple shows there on some of those event driven ones that it could very well be audio only, but and those will be in your usual suspects, Apple and Google, Amazon, all the standard distribution, Spotify, et cetera.

So with that I'm going to sign off today. Look for a new episode once a week on Fridays. This episode is an introduction to the show, of course, but hopefully you'll get excited about this and see how much is possible. I do think there's a huge opportunity for businesses to use AI across a broad set of use cases. Over the next few weeks and months, we'll look at some of those. Thank you very much for listening to this episode zero, this first episode. And I'm looking forward to seeing you again on the next episode where we'll have some interesting topics to talk about AI.

Thanks for joining me.”