Although the use of artificial intelligence has steadily increased over the past decade, the recent release of OpenAI’s generative AI system, ChatGPT, has resulted in a precipitous rise in the attention and publicity that accompanies the rise of powerful generative AI systems.
These generative AI systems are increasingly accompanied by problems and concerns about the use of AI systems by technology service providers.
What is an AI system?
To demonstrate how these generative AI systems work, we asked ChatGPT to answer the question “What is an AI system and how does it work?”. His answer:
An AI system is a computer program that can perform tasks that normally require human intelligence, such as understanding natural language, recognizing objects in images, or making decisions. Generative AI is a type of artificial intelligence that uses algorithms to generate new data, such as images, text or audio, that has never been seen before.
If we were to submit this query to a general search engine, we could get responses containing the terms “AI system” and would then review each resource and compose a response based on the information provided. The difference between performing a general search and using generative AI is that generative AI assembles the answer for you, complete with correct syntax and wording in response to the question.
How does an AI system work?
AI systems analyze large amounts of data and extract patterns and insights from it. The analytics system can learn to recognize patterns in data and use those patterns to make predictions or classifications about new data. In other words, artificial intelligence is basically an incredibly complex and immensely large decision tree based on statistical probability.
Technically, these decision trees take the form of Markov chains or stochastic methods. A Markov chain is a mathematical model that describes a system that transitions between different states over time. It is a stochastic process, which means that the system’s next state is determined by a probability distribution that depends only on the system’s current state and not on previous states. In natural language processing, Markov models are often used to model the probability distribution of words in a text. For example, a first-order Markov model would predict the likelihood of a word based on the likelihood of the previous word. A second-order Markov model would predict the likelihood of a word based on the likelihood of the previous two words.
Markov chains provide a powerful framework for modeling and understanding sequential data in machine learning and AI applications. However, it is important to note that AI systems require a significant amount of data to “train” the algorithm, or derive the probabilities needed to build the chain. In the case of a language model like ChatGPT, the application has been trained on huge amounts of text data and has learned to produce natural-sounding speech by predicting the most likely next word or phrase based on the previous context. ChatGPT advertises that it gets its material from a variety of sources and domains, including books and literature, websites and articles, as well as social media and messaging. Much like training a dog, an AI system must go through repeated patterns in order to “learn” to deliver the desired result. The source of this training data can be a hot topic and worth considering as AI becomes more important.
What are some examples of an AI system?
We’ve become accustomed to AI systems like Siri and Alexa, each of which can be fed requests and queries, and in turn can select responses and complete tasks from a closed list of possible responses. Other well-known AI applications include the text-completion feature in certain email products, image recognition in popular social media sites that suggests location or personal tags, and autonomous driving or self-driving mechanisms in cars. And of course, we’ve become accustomed to AI chatbots in a number of contexts.
Unlike previous AI bots, ChatGPT’s replies remember the context of the ongoing conversation and can be prompted to perform deeper or secondary instructions, such as: B. giving an answer that fits a certain style or using certain defined terms. These generative capabilities have enabled a number of new players in the technology scene around AI, but some better-known service providers are also developing competitive AI systems.
As these AI systems become more prevalent in business environments, the reality is that using this technology is not without risk. In Part 2 of this series, we examine the legal risks of AI systems in technology services.