Montgomery AI Notes and Thoughts

On March 5th, the Montgomery AI event was held at Alabama State University (ASU). There were a variety of speakers there. All of them were from the local area or nearby areas. The objective of the event, was to discuss the various aspects of AI (Artifical Intelligence) including, how to use it, the best way to write prompts, and how to use agents to improve the language models.

Overall it was informative. One thing I did not see was people from the Alabama Technology Quality Assurance Board. This board was created in part, based on the work done by the Generative AI Task Force that was created in years prior.

Session 1B

• open code; open source tool
• command line ai agent
• follow instructions on https://opencode.ai/docs for how to install on your system

Session 7B

• agent takes action based on what you ask it to do
• chat bot and agent are not the same; agent can look online or other sources for information; the chat bot only knows what it * knows and can only answer based on what it knows;
• consider looking at custom GPT
• can create custom agent using ChatGPT. requires paid subscription
• consider using canvas for essays. allows for inline editing and make adjustments to the previously provided output
• can make a consel of LLM or AI. then they consult and advise on something or different way of doing research
• n8n, ai automation workflow

Session 8B

• focus on prompting
• prompt to product
• be mindful of the guardrails and not put in identifiable informatino
• one of the essential features is to be through

Other Session Notes

• City creating their private AI that keeps information contained and separate
• City does not allow their AI to pull information from the internet. it can only utilize internal docs
• Submit suggestions about changes to be made to the Open Data portal, so that it can be improved