The ITRAK AI Journey
Executive Summary
In today’s rapidly evolving technological landscape, it has become very difficult to discern how and when to invest in what is to come. The wave of change new tech brings can be incredibly uncertain and confusing but also integral to the longevity of a business. Leveraging the power of artificial intelligence (AI), ITRAK has embarked on an exciting journey, exploring various AI models and crafting innovative Health & Safety use cases. In this blog post, we will provide a high-level overview of ITRAK’s AI journey, highlighting our successful experimentation with different models and shedding light on our future roadmap.
The Chronology of AI Model Experimentation:
With the advent of the GPT3 model, many have taken up a new perspective on the usage of AI in their personal daily activities and larger business processes. ITRAK’s focus has been leveraging the GPT3 and other models for Operational System Support (OSS). This approach targets the usage of these models for use cases that promote efficient, effective and safe workflows and processes.
OpenAI Whisper Model: Empowering Voice Assistance
ITRAK’s first foray into AI commenced with the utilization of the OpenAI Whisper model. This cutting-edge model has been instrumental in enabling ITRAK to experiment with text-to-speech and speech-to-text voice assistance. By harnessing the power of the Whisper model, ITRAK has achieved remarkable success in capturing user input effectively and delivering audible, informative messages. This breakthrough has paved the way for enhanced user experiences and streamlined communication channels within ITRAK’s ecosystem.
OpenAI Text Davinci and GPT 3.5 Turbo Models: Unleashing the Power of Text
Continuing their AI exploration, ITRAK delved into the realm of text-based AI models, leveraging OpenAI’s Davinci and GPT 3.5 Turbo models. These models have been employed across diverse use cases within the organization. Notably, ITRAK has utilized the GPT 3.5 Turbo model to capture unstructured data through user input and transform it into structured formats, such as generating corrective action drafts. Furthermore, these models have been instrumental in developing an internal OSS that empowers ITRAK users to efficiently locate the information they need. Looking ahead, ITRAK aims to harness these models as a recommendation system, facilitating users in completing forms comprehensively and concisely.
Exploring AI Competitors and Alternatives:
Driven by the pursuit of excellence, ITRAK has actively tested various competitors and alternatives to GPT models, such as EluetherAI’s GPT-neo-2.7B, pythia-6.9b-deduped, and DialoGPT-medium. However, due to considerations around speed and performance, these alternatives yielded unsatisfactory results. ITRAK remains committed to staying at the forefront of AI advancements, continuously seeking innovative solutions to elevate their AI capabilities.
Langchain: Unlocking Conversational Agent Potential
Continuing their relentless pursuit of innovation, ITRAK recently ventured into the realm of Langchain, the newest LLM framework. This framework empowers AI engineers to build conversational agents using custom-designed modules tailored for data pipelines and other conversational use cases. Excitingly, ITRAK’s team successfully leveraged Langchain’s modules to generate summary documents for 65 webinars in under a minute, revolutionizing their workflow efficiency. With ongoing experimentation on other modules, such as Langchain tools, ITRAK aims to equip their AI agents with simple capabilities like scheduling meetings, further streamlining their operations.
Future Outlook: Transforming EHS Workflows
As a pioneering leader in the EHS industry, ITRAK remains committed to responsibly and effectively leveraging technology to transform traditional and cumbersome processes. Keep a close eye on ITRAK’s upcoming AI tools, which will revolutionize procedures, hazard assessments, and risk scoring, enabling smoother and more valuable workflows.
Conclusion
ITRAK aims to revolutionize traditional processes and redefine the landscape of EHS workflows to unlock the real human value.
Join us on this journey to revolutionize EHS workflows, enhance procedures, and make hazard assessments and risk scoring smoother and more valuable than ever before.
Stay tuned for our upcoming AI tools that will reshape the landscape of the EHS industry. Let’s unlock the true human value together!