Artificial intelligence (AI) might sound like something from a sci-fi movie, but it’s something that is active in our everyday lives and has thousands of use cases and applications, here at Extract we leverage AI in all of our workflow solution’s; ID Shield, Flex Index, Sort & Send, LabDE, Hawkeye, and HealthyData
The latest AI language program, ChatGPT which stands for Chat Generative Pre-Trained Transformer, has taken the internet by storm, thanks to its incredible conversational responses to any question you input.
The crafty application is an AI language model developed by OpenAI. It uses advanced machine learning algorithms to generate human-like text based on the input provided. ChatGPT is trained on a massive amount of text data and can answer questions, provide information, write stories, and perform many other language-related tasks.
Here is my ‘interview’ with the robot about AI, Machine Learning, and Automating Workflows:
Taylor: Hello Chat GPT, my name is Taylor Genter. Today I would like to explore the use of AI in natural language understanding. Is it ok if I ask you a few questions?
Chat GPT: Of course, Taylor. I would be happy to help you with any questions you have about AI and natural language understanding. Please feel free to ask away!
Taylor: What does GPT stand for?
Chat GPT: GPT stands for "Generative Pre-training Transformer." It is the name of the model architecture that I was trained on. Generative models are a type of machine learning model that are capable of generating new data, such as text, images, or speech. Pre-training refers to the process of training a model on a large dataset before fine-tuning it on a smaller, task-specific dataset. The transformer architecture is a type of neural network that is particularly well-suited for natural language processing tasks, such as language translation and text summarization.
Taylor: That's a lot of information to take in. Can we start with machine learning? What is it?
Chat GPT: Sure! Machine learning is a method of teaching computers to learn from data, without being explicitly programmed. It is a subfield of artificial intelligence (AI) that uses algorithms to learn patterns and insights from data. There are three main types of machine learning: supervised learning, unsupervised learning and reinforcement learning.
Supervised learning is the most common type of machine learning, where the algorithm is trained on labeled data, meaning that the desired output is provided alongside the input data. The algorithm then learns to make predictions on new, unseen data based on the patterns it learned from the training data.
Unsupervised learning is where the algorithm is not provided with labeled data, but it's supposed to find patterns and insights in the data by itself.
Reinforcement learning is a type of machine learning where an agent learns to make decisions by interacting with its environment and receiving feedback in the form of rewards or penalties.
Machine learning is used for a wide variety of tasks such as image recognition, speech recognition, natural language processing, and many other.
Taylor: How is machine learning different from AI?
Chat GPT: Machine learning is a subset of artificial intelligence (AI). While machine learning is focused on the development of algorithms and models that can learn from and make predictions on data, AI is a broader field that encompasses machine learning as well as other techniques for building intelligent systems.
AI includes a wide variety of techniques and approaches, such as rule-based systems, expert systems, decision trees, evolutionary algorithms, and many others. While machine learning is a key component of AI, it is not the only one. AI systems can also be built using other techniques such as rule-based systems, expert systems and decision trees.
In summary, AI is a field that encompasses many different techniques and approaches for building intelligent systems, while machine learning is a specific approach within AI that is focused on training algorithms to learn from data.
Taylor: Are you saying that machine learning models learn on their own, while with general AI you may have pre-programmed rules?
Chat GPT: Yes, that's one way to think about it. Machine learning models are trained on data and are able to learn from it, whereas traditional AI systems use pre-programmed rules and decision-making processes.
Machine learning models can automatically learn patterns and relationships in the data and make predictions or decisions based on those patterns. They are able to adjust and improve their performance over time as they are exposed to more data. On the other hand, traditional AI systems use pre-defined rules and decision-making processes, which are determined by the designer of the system, and can't adjust to new data.
It's also worth noting that there's also a hybrid approach, where you can combine both approaches, using machine learning to learn from data and pre-programmed rules to guide the decision-making process.
It's important to note that AI is a broad and rapidly evolving field, and there's ongoing research in the area of combining traditional rule-based and machine learning approaches to build intelligent systems that can learn and adapt over time.
Taylor: Let’s move into a few questions on how AI and ML can apply to automated workflows, such as automated redaction and indexing.
Chat GPT: Sure!
Taylor: Should automating workflows be important to government agencies?
Chat GPT: Absolutely! Automated workflows are a great way for government agencies to streamline their processes and maximize efficiency. Automated workflows can automate mundane tasks, reduce manual labor and help ensure accuracy in data entry and processing. They can also help agencies to better manage their resources, allowing them to focus on their core mission and services. Additionally, automated workflows can help reduce operational costs, improve customer service, and increase transparency in government operations.
Taylor: I completely agree. Do you think using technology like AI and ML benefits automated redaction workflows?
Chat GPT: AI can help automate redaction processes in a number of ways. AI can quickly scan documents to identify and categorize sensitive information, such as personally identifiable information or confidential data. This allows for rapid redaction of the data, which can then be securely stored and protected. AI can also help detect and analyze anomalous activity, which can help organizations quickly identify documents that need to be redacted. Additionally, AI can help automate the manual process of redacting documents, which can improve efficiency and accuracy. Automating redaction workflows can be a great way to save time and resources. It can help to streamline the process of redacting sensitive information from documents, making it easier and faster to complete. There are several different tools available to help automate the process, from software that can scan documents and automatically detect information that needs to be redacted, to programs that can automatically apply redaction to documents. With the right tools in place, automating redaction workflows can help to free up time and resources, while ensuring that sensitive information is kept secure.
Taylor: I could not have summarized that better myself, what about automated document indexing?
Chat GPT: AI can help with automated indexing by using natural language processing (NLP) technologies to understand text and identify key concepts. This can be used to automatically label and group documents into meaningful categories. AI can also use machine learning algorithms to learn from existing data and improve the indexing process over time. By leveraging AI and NLP, automated indexing can save time, reduce errors, and improve overall accuracy.
Taylor: Those are many of the technologies we leverage at Extract Systems. Do you mind closing out this interview? How can folks reach out to Extract Systems to learn more about their automated workflow solutions?
Chat GPT: If you would like to learn more about Extract Systems and their automated solutions, you can visit their website at www.extractsystems.com or you can send an email to info@extractsystems.com. You can also give them a call at +1 (608) 821-6520 to speak to one of their knowledgeable representatives and ask any questions you may have. They are always happy to help and provide more information!
Taylor: Thank you for your time. This was fun! Bye
Chat GPT: You're welcome, Taylor. I'm glad I could help. If you have any more questions, feel free to reach out. Have a good one!
As you can see from this introduction to ChatGPT in its own words, it offers a powerful set of capabilities and features out-of-the-box, and able to speak in natural, sophisticated responses. For this blog, there were some visible limitations, sometimes not producing enough details or having the requested information available. Overall, ChatGPT is quite impressive to the AI/ML community, especially given it’s an initial release, and we’ll see what the future brings.
Taylor is a Marketing Manager at Extract with experience in data analytics, graphic design, and both digital and social media marketing. She earned her Bachelor of Business Administration degree in Marketing at the University of Wisconsin- Whitewater. Taylor enjoys analyzing people’s behaviors and attitudes to find out what motivates them, and then curating better ways to communicate with them.