Undoubtedly you’ve read everything from headlines to deep-dive thought pieces on the promises and perils of AI for every corner of every industry. Here are three quick take-aways about why it’ll change how you work, but not do the work itself.

1. Talking to machines is not as immediately easy or valuable as you think.

While AI has likely disrupted portions of every workflow, it has also added a critical skillset very few have yet mastered: talking to it. Receiving an intelligent and usable answer from the AI system is directly related to one’s ability to generate an intelligent prompt for the system.

Think of it this way: asking for “a cake recipe” is not the same as asking for “a 19th century baking recipe, adapted for vegans who live in Northern Europe today who have access to limited agriculture, which can be prepared in under 60 minutes, and will serve 8 people.”

2. Expert work requires expert knowledge and expert experience.

Expert output from an AI system is ideally easy to digest and implement, quickly and efficiently answered, and being trustworthy and accurate. To achieve this, you need an expert level of knowledge so you can ask the more specific questions further down the rabbit hole. It’s one thing to ask an AI system how the stock market works, it’s another thing to ask it for modelling data for a specific type of financial product in a specific market context.

Finally, one’s expert experience must be applied to deduce how practical, usable and trustworthy the AI’s answer is, not least because large language models like ChatGPT aren’t always accurate. In other words, if you don’t already know a ton about Art Nouveau, no one’s going to hire you to help them solve questions around Art Nouveau, AI or not.

3. Embrace your “inner Alice” and go down the rabbit hole.

AI can be used for contextual search that is conversational in nature, allowing you to ask follow-up questions and dive deeper into a topic. Traditional search engines like Google typically retrieve information based on keywords, but they don't have the ability to understand the context of the search or maintain a conversation. Once you've received the results, you need to manually sift through the information and formulate a new search query if you want additional information.

AI-powered search, on the other hand, can understand the context of the conversation and provide answers tailored to the user's specific needs. For example, if you ask an AI-powered search engine about a topic, such as "Alice and the rabbit hole," it can provide an answer and then you can ask follow-up questions, such as "What happened next?" or "Who was the Cheshire Cat?" The AI system can then use this information to provide additional, more specific answers, allowing you to delve deeper into the topic.

In this way, AI-powered search can simulate a conversational experience, letting you ask follow-up questions and get more targeted, relevant results. This can make the search process more efficient and enjoyable, since you can get the information you need without having to manually sift through a large amount of irrelevant information. Oh, and case in point: this answer (and only this answer) was generated amidst a 7-layer deep AI conversation once it truly understood the concept and metaphor I wanted to use.

What next?

You probably have questions and, likely, ChatGPT hasn’t given them to you yet. We’re always around to talk through how AGM is integrating AI across all services and solutions globally, and how that might be relevant to you and your business. Get in touch HERE.