Skip to main content

What possible use could Gen AI be to me? (Part 1)

There’s a great scene in the Simpsons where the Monorail salesman comes to town and everyone (except Lisa of course) is quickly entranced by Monorail fever…

He has an answer for every question and guess what? The Monorail will solve all the problems… somehow. The hype around Generative AI can seem a bit like that, and like Monorail-guy the sales-guy’s assure you Gen AI will solve all your problems - but can be pretty vague on the “how” part of the answer.


So I’m going to provide a few short guides into how Generative (& other forms of AI) Artificial Intelligence can help you and your team. I’ll pitch the technical level differently for each one, and we’ll start with something fairly not technical: Custom Chatbots.


ChatBots these days have evolved from the crude web sales tools of ten years ago, designed to hoover up leads for the sales team.


They can now provide informative answers to questions based on documents or websites. If we take the most famous: Chat GPT 4. If we ignore the standard free version, that every other student is using to churn out essays, and look at the paid-for version - that will be a good entry point.


This $20 a month service allows you to create and distribute custom Chat GPT’s. This will have all of the conversational ability of the latest chat GPT, but you can also get it to base answers on documents you give it. You can also tailor how it responds, for example should it be casual or formal, provide examples or just “answer the question” only. Use college level language or that of child e.g. “explain it to me like I’m 5”


Behind the scenes, Chat GPT is doing a semantic search of your documents, that is - search based on the concepts of your question - rather than a crude keyword search. This approach known as RAG (Retrieval Augmented Generation - fancy name for a simple thing) allows Chat GPT to look for information on the concepts held in your query - and then answer using those concepts, woven into a single coherent response. Ie: it retrieves data using a clever search and uses that data to augment the answer it generates.


A helpful robot business analyst
AI Chat Bots can be an invaluable aid to learning and delivery


You can actually go further and get Chat GPT to base its answers based on an API call into your systems. This API call might do a calculation (ChatGPT can struggle with math questions) or query your database. It might also set up a calendar invite, score the results of a game or find cheap flights - if you connect it to something like SkyScanner.


So how do I use these custom AI chat-bots? I currently work in the field of SWIFT Payments. (Ever seen a movie where they “wire” the money round the world - they are using SWIFT payments.) These payments are based on complex messages and are heavily regulated by central banks and governments. As you can imagine - in this sweet spot between Government regulation, banks and complexity lives a lot of long dry documentation. 


So I took these publicly available documents and specifications and put them into a Chat GPT chat bot - along with a few sentences on how to discuss and explain the answers to my questions. The end result is a very helpful and knowledgeable assistant that can answer questions varying from what are the options for a data field in a complex payment message up to what is the regulatory impact of system X or Y being deployed in the cloud. 


Of course this information can be found in Google, but the difference is - Google will correctly give me a link to a 400 page PDF that contains my answer. Correct but not useful.   Meanwhile - My custom Chat GPT chatbot will essentially read that document for me, and answer based on what it found - That's useful.


Modern AI chatbots can do much more than this - the above is possible in 10 minutes, with guidance and a willingness to experiment - you can build AI tools that help you and your team learn, adapt and deliver faster.

Labels:

Comments

Post a Comment

Popular posts from this blog

Can Gen-AI understand Payments?

When it comes to rolling out updates to large complex banking systems, things can get messy quickly. Of course, the holy grail is to have each subsystem work well independently and to do some form of Pact or contract testing – reducing the complex and painful integration work. But nonetheless – at some point you are going to need to see if the dog and the pony can do their show together – and its generally better to do that in a way that doesn’t make millions of pounds of transactions fail – in a highly public manner, in production.  (This post is based on my recent lightning talk at  PyData London ) For the last few years, I’ve worked in the world of high value, real time and cross border payments, And one of the sticking points in bank [software] integration is message generation. A lot of time is spent dreaming up and creating those messages, then maintaining what you have just built. The world of payments runs on messages, these days they are often XML messages – and they can be pa
Post a Comment
Read more

Is your ChatBot actually using your data?

 In 316 AD Emperor Constantine issued a new coin,  there's nothing too unique about that in itself. But this coin is significant due to its pagan/roman religious symbols. Why is this odd? Constantine had converted himself, and probably with little consultation -  his empire to Christianity, years before. Yet the coin shows the emperor and the (pagan) sun god Sol.  Looks Legit! While this seems out of place, to us (1700 years later), it's not entirely surprising. Constantine and his people had followed different, older gods for centuries. The people would have been raised and taught the old pagan stories, and when presented with a new narrative it's not surprising they borrowed from and felt comfortable with both. I've seen much the same behaviour with Large Language Models (LLMs) like ChatGPT. You can provide them with fresh new data, from your own documents, but what's to stop it from listening to its old training instead?  You could spend a lot of time collating,
Post a Comment
Read more