iCreatAI News
Skip to content
Trending storyWatching

Using ontology plus validated models to let AI agents give traceable pricing answers

1 reports1 sources2 days agoUpdated

Understand this story

AI overview

Safdar Ali shares a transparent pricing approach for AI agents: organizing domain knowledge with an ontology, paired with validated predictive models, invoked by AI agents to give traceable answers without fabricating content. The approach is based on practice in the Australian aged care industry, used to answer accommodation pricing questions. Under local regulations, accommodation prices above A$789,686 require approval by a government agency and must be supported by evidence-based justification. Ali says the relevant answers have helped aged care providers meet this evidentiary requirement. On implementation requirements, Ali says training a model covering more than 9,000 rooms in Australia can be done with CPU alone; however, running the AI agent may require access to third-party platforms offering a zero data retention (ZDR) option, such as AWS Bedrock, OpenRouter and Fireworks AI. For maximum data privacy, one could consider running locally on a mini PC with a commercial-grade GPU (such as Nvidia DGX Spark or Apple Mac Studio), which cost around A$9,500.

AI overview based on source reports · Updated 2 days ago

Coverage timeline

Follow the coverage to explore different perspectives.

Nov 16
  1. Hacker News AI
    Ontology in the Age of AI

    The author describes building an ontology and validated prediction model for Australian aged care accommodation pricing, which an AI agent then uses to answer regulator questions without fabricating data, in a sector where room prices above A$789,686 require government approval backed by evidence. The piece covers ontology definitions from Gruber (1993) and Noy & McGuinness (2001), a worked retail example, the aged care ontology design journey, and cautions about black-box predictions, data leakage on third-party LLM platforms, and token-based costs. It concludes with takeaways including that LLMs can be swapped but the ontology and validated model cannot, and that ontology should be designed before the AI agent is developed.

Activity for this story

Current comparable activity
3
Peak comparable activity
10Nov 16 21:00
Change in the past 24 hours
-51%

Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.