Artificial intelligence (AI) has been making headlines recently. Whether you’re making
‍questionable crochet narwhals or generating hip-hop lyrics about meat, it’s clear that the latest generation of AI tools can provide a lot of entertainment.  

‍

But is AI valuable to your business? What AI tools exist these days,  and should you be adopting them yet?  

‍

Keep reading to learn more about what artificial intelligence is and whether you should be using it in your organization.  

What is artificial intelligence?  

Artificial intelligence, or AI, is when software is programmed to perform tasks that typically require human effort. Examples of these tasks include detecting patterns in data, analyzing complex financial models, and developing new products. AI is capable of performing such a wide range of tasks because it can learn and adapt to new information over time — as it gathers more data, it becomes better at performing its assigned tasks.  

‍

AI is designed to imitate human behaviors and capabilities; iit even wrote the paragraph above (this wouldn’t be a proper intro to AI without that bit 😏). 

‍

Long before artificial intelligence was tweaking our social media feeds and serving up Google search results, books and films familiarized us with the idea of “intelligent” anthropomorphic machines that could think like humans. That probably contributes to some misconceptions (and unease) around AI. So let’s move away from vague terms and talk about some of the specific capabilities modern artificial intelligence has.  

Machine learning 

Machine learning is the way we train machines to recognize patterns in data and draw conclusions and predictions from those patterns. It often lays the foundation for an AI system.  

Anomaly detection 

Anomaly detection is the ability to automatically detect errors or data points that deviate significantly from the rest of the data. For example, it’s used to detect things like fraudulent credit card activity or tumors on an MRI image.  

Computer vision 

Computer vision allows software to recognize and interpret images and videos. For example, computer vision is key to self-driving vehicles, which use it to understand things like traffic lights, signs, and other vehicles on the road.  

Natural language processing 

Natural language processing gives computers the ability to interpret written or spoken language and respond in kind. You’ve probably already interacted with natural language processing systems a lot, like when you use predictive text on your phone or get a search result that matches your intent but not your exact wording.  

Knowledge mining 

Knowledge mining involves extracting information from large, often unstructured sets of data. It can index the data for easy searches and provide insights.    

‍

What is generative AI? 

Artificial intelligence is nothing new, so why have we been hearing about it so much lately? 

‍

Most of the news is about recent breakthroughs in generative AI. Generative AI is artificial intelligence that can learn from input to produce new content. For example, if it’s trained on examples of code, it can turn around and create its own unique code that’s (in theory, at least!) semantically and syntactically correct.  

‍

OpenAI has released a new set of generative AI apps that have taken the world by storm. ChatGPT can generate text that’s almost indistinguishable from human writing, whereas Dall-E takes text input and creates images (which are impressive, but a bit weirder if we’re honest).  

‍

Let’s take a look at how generative AI fits into the artificial intelligence landscape. 

‍

At the highest level, we have artificial intelligence, or software that imitates human behavior. Machine learning, as we’ve already discussed, is a subset of AI. Machine learning models take in data and fit the data to an algorithm in order to make predictions.  

‍

Next, we have deep learning, which is a type of machine learning based on artificial neural networks. These neural networks make sense of observational data by passing it through interconnected layers of nodes, generating a more complex understanding with each one. Deep learning is particularly effective for tasks such as image recognition and natural language processing.  

‍

Finally, generative AI is a subset of deep learning. While some types of deep learning focus on classifying items or making predictions, generative AI is about creating brand new content.  

‍

Illustration of generative AI's place within machine learning. 

(Source: Microsoft AI Course) 

‍

Want to learn more about AI and its application in business? Stay tuned for Part II, where I’ll dive deeper into AI in 2023 and beyond, as well as using capabilities within your organization to stay ahead.

‍

About Tom Hart

Hi, I'm Tom, and I'm a curious connector ⚡🤝 By combining an abundance of curiosity with 9+ years of experience in software development, video production, and solutions roles, I am able to generate traction in dynamic environments. Essentially, I help create shared understanding and cross-functional alignment by asking questions and then synthesizing the resulting answers into proposals and next actions.