Beyond Complex AI: How Kane Hooper Makes It Work for Your Business
From Personal Struggle to Universal Solution
Kane's ideas and approach to AI have the potential to significantly change how customer service is delivered.
At the RubyConf AU 2024, Kane, CEO of reinteractive and a developer with a mission, took the stage to share a deeply personal story. His wife's medical condition robbed her of her voice, leaving them searching for ways for her to communicate again. This experience sparked a fire within Kane, leading him to develop an application that utilised AI to clone a person's voice for phone calls.
This anecdote wasn't just a heart-wrenching tale; it served as a powerful introduction to Kane's core message: AI's potential to improve lives goes far beyond technical prowess. And with that, he dove into his groundbreaking approach to making AI accessible for everyone, specifically focusing on revolutionising customer service.
Introduction
Kane's Audio Demo
Kane demontrates how he used a 1-minute audio clip of his wife's voice to create a voice for her using llElevenLabs.
AI for Everyone: No More Black Boxes
Kane emphasised the recent shift in AI – it's no longer shrouded in mystery. Pre-built models and user-friendly APIs have made AI development more approachable than ever. He demonstrated how a simple API call can be used to clone a voice and then another to make a phone call with that cloned voice.
Demo on how a simple API call can be used to create a cloned voice
Demo on how to make a phone call with a cloned voice.
The Dark Side and the Bright Light
Kane acknowledged the potential dangers of AI, highlighting the possibility of creating deepfakes – manipulated videos that can have malicious purposes. However, his focus remained firmly on the positive. He championed open-source AI models, which offer a transparent and cost-effective way for businesses to leverage AI's power.
The Power of Prompts and Knowledge Retrieval
Kane delved deeper into the technical aspects, explaining the importance of "prompting." In essence, prompts are instructions that guide AI models towards the desired output. He also discussed "knowledge retrieval," where AI accesses and processes information from various sources to deliver the most relevant response.
The importance of prompting.
Knowledge retrieval
For businesses, imagine a system that can instantly search a company's data troves based on a customer's query. The AI fetches relevant information, empowering a company to deliver faster, more precise answers.
Vector Embeddings: Making Sense of Information
For those curious about the technical underpinnings, Kane introduced the concept of vector embeddings. This technique translates information into a numerical format, enabling AI to perform similarity searches with incredible efficiency.
From Theory to Reality: Open Source Takes Center Stage
Kane wasn't just about theory. He showcased the power of open-source AI with real-world applications. One example involved a client in the medical field. AI helped navigate the maze of patient data – emails, reports, and medical records – saving doctors countless hours searching for critical information.
Fine-Tuning: AI Learns Your Business
But what about customisation? Here's where Kane introduced "fine-tuning." These open-source models can be trained on a company's specific data, enabling them to tackle unique challenges. Imagine an AI system adept at processing complex financial documents or navigating industry-specific regulations.
These open-source models can be tweaked to address a company's specific needs. With a relatively small set of training data, the AI can be fine-tuned to excel in a particular domain, with more examples like processing insurance claims or navigating banking regulations.
Ollama: Your Open Source AI Playground
For those eager to experiment with AI, Kane introduced Ollama, a user-friendly tool that allows anyone to download and run open-source models locally. This removes technical hurdles and empowers businesses to explore the possibilities of AI without significant investment.
The beauty of Kane's approach lies in its openness. Forget hefty licensing fees – he champions open-source AI models, readily available and free to use. This removes a significant barrier for businesses hesitant to embrace AI due to cost concerns.
Simplifying the Process: AWS Bedrock to the Rescue
For those seeking an even more streamlined experience, Kane mentioned AWS Bedrock. This service acts as a one-stop shop for interacting with various AI models through a single API call. No need to write complex code – just a simple API call unlocks the power of different AI functionalities.
A Glimpse into the Future: The AI-Powered Call Center
Kane didn't just talk about the future, he showed it. He presented a live demo of a call center system he built using AI. The system intelligently routed calls to the appropriate agent based on the caller's initial inquiry. This glimpse into the future showcased the potential of AI to transform customer service into a faster, more accurate, and ultimately, more satisfying experience for everyone.
AI for Human Good
Kane concluded by emphasising the potential of AI for positive social change. He envisions a future where AI complements human interaction, freeing us from tedious tasks and allowing us to focus on what we do best – building meaningful connections. The message was clear: AI is not a replacement for human interaction, but a powerful tool to elevate it.
Inspired by Kane's vision?
Free Consultation
Kane is offering consultations to help businesses explore how AI can be integrated into their applications and projects. Whether you're a seasoned developer or just starting your AI journey, Kane's open invitation offers a valuable opportunity to learn and unlock the potential of AI in your customer service experience.
