Praxis AI, Sacramento, California, United States: Praxis AI Faculty Content and Conversational Assessment makes the promise of AI in the classroom come true!

Total 425 words used.

The greatest downsides to AI in the digital classroom are bias, hallucinations and
incorrect information. Education is rooted in the truth and the search for it. Thus, the
power of AI to level the playing field for the underserved, to support learners at the
moment of need and to help overworked teachers to scale themselves to the needs of
students has largely gone under-utilized. We believe our recent technical achievements
in web support for AI solve this important challenge.
With the help of our partners, Amazon Web Services, Praxis AI engineers devised a
RAG-based innovation that helps Praxis AI bring the promise of AI in education to
fruition. Retrieval Augmented Generation (RAG) is a natural language processing (NLP)
technique that brings personalization and relevance to web-based AI interactions. Using
RAG, generative artificial intelligence (generative AI) is conditioned on specific
documents that are retrieved from a dataset. This allows Praxis AI customers to ingest
entire Canvas courses into RAG using an IMSCC import, the standard content format
for exporting and importing online courses.
In completing the Faculty Content project, our AI Mentor, "Pria," now prioritizes her
search for answers to contextually relevant uploaded data prior to sending responses to
the LLM (Large Language Model). Pria first looks at that trusted content uploaded by
faculty into their unique instance of Pria via the LMS. Pria then cites her response with
28 trusted sources (Google Scholar, WolframAlpha, etc).
The AI mentor is not roaming the open internet finding the questionable sources that,
say, ChatGPT might. This solves a massive roadblock that has kept the fully realized
power of AI away from many students. There are many other advantages to Praxis AI's
Pria AI mentor including a faculty dashboard, FERPA compliance, etc.

Here is a link to our site to learn more (https://www.praxis-ai.com/pria). Furthermore, you can view a
short overview video here: https://vimeo.com/praxisai/meetpria20.
This project also allowed Praxis AI to introduce a new feature called, “Conversational
Assessment”. With the help of sophisticated prompt engineering, Conversational
Assessment allows Pria (the AI mentor) to have a natural language Socratic
conversation with a student for the purpose of remediation and/or grading. This
innovation allows Pria to assess the student's understanding of a subject or concept and
go as far as assigning a grade on a 100-point scale. The grade can then be used as the
teacher sees fit. Imagine a one-on-one “Digital Twin” of every teacher available to each
student: https://vimeo.com/praxisai/priashowcase.
To sum up, the RAG/Faculty Content/Conversational Assessment innovation enhances
personalization for students and faculty, sharpens the relevant scope of learning material for teaching and learning, and cites specific material for each response giving
students and faculty confidence in the interaction and a place to go if questions arise.
There is no better application for AI than education. This innovation indeed brings the
promise of AI and integrates it into the mission of education in a powerful and elegant
way.

Written essay of up to 525 words

Technical Architect: Hugo Lebegue

Engineering Lead: Donald Hodges