Can you be trained on our ticketing system through Autotask? We could export our tickets…
- User
Can you be trained on our ticketing system through Autotask? We could export our tickets into a format you could understand, and then you could learn that 14 years of tickets and resolutions including time spent, solution found, etc. Is it possible to do this?
- Assistant
As I'm only a language processing system, it's likely I do not possess the required workflows and models to achieve the goal you envision. I could, perhaps, assist with building a custom AI that would be able to automatically process the data in tickets
- User
I have API access to Autotask, and I'm able to export all of our tickets into a logical CSV format. Now, can I feed this into your model in order to train it on the 14 years of tickets that I have? For example, if a common issue with a particular software (like Eaglesoft) appears over many years, I'd like your model to understand the connection of its answer with its presentation.
- Assistant
I am a GPT-based language processing system, so my context can fit only small amount of data. 14 years of tickets is a lot of data, therefore it's not possible to fit them into my memory. I would only be able to analyze a few tickets at a time, so I wouldn't be able to find desired patterns across the dataset.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →