The Rise and Reign of Generative AI: Everyone, Every Industry
Let’s cut the jargon and dive into what’s cooking in the world of generative AI. So, you’ve got this keynote from Swami Sivasubramanian, Amazon’s VP of the tongue-twister department—Database, Analytics, and Machine Learning. Mr. Swami, foretelling like Nostradamus, reckons that AWS’s superpowers will thrust generative AI into the hands of all, no matter the business line. Engineering, marketing, customer service, finance, sales, you name it.
“Generative AI has captured our imaginations,” he says. Why? Because it’s a turning point for machine learning. It’s these mind-boggling large language models (LLMs), pre-trained on a gargantuan amount of data, and cleverly known as “foundation models” (FMs). It’s like if Yoda of Star Wars fame turned to the dark side and decided to dominate the world of AI.
Now, here’s the takeaway: generative AI is set to upgrade your experiences, whether you’re interacting with virtual assistants, enjoying personalized shopping, or even working on mundane office tasks. Imagine seeing your productivity skyrocket with conversational search, or text summarization, or maybe, a code generation tool that isn’t going to rebel against you on a moody Monday.
In terms of business operations, it’s all about the intelligence—document processing gets smarter, quality controls become sharper. But how, you ask? AWS, says Mr. Swami, is set to be the metaphorical key that unlocks this treasure trove of generative AI.
The ABCs of AI: A Generative Journey
No one model will take the Iron Throne. Organizations will pick the right model for the right job. Your ad company may want to tweak a model with top performing ad copy, while a retailer could offer the model access to inventory data. It’s all about customization, and it’s all under the safe umbrella of AWS. And let’s not forget the Generative AI Innovation Center, where customers get the chance to build and deploy custom AI products and services.
What’s the formula for AI democratization within organizations? It’s easy-to-use tools plus low cost and low latency responses. Thanks to purpose-built ML infrastructure, this is no longer a pipe dream. Much of this innovation is expected to ride on the coattails of Amazon Bedrock—a service by AWS that allows organizations of any size to build and scale their own AI applications.
For the past quarter of a century, Amazon has been dipping its toes into AI and ML tech. Their recent innovations have now made the capabilities of generative AI not just a possibility, but a reality. So, what’s new?
The Amazon Way: Seven Generative AI Updates
AWS is pumping up Amazon Bedrock with new model providers and additional FMs. For example, they’re welcoming Cohere’s text generation model, Command, and Anthropic’s latest language model, Claude 2. They’re also releasing the latest version of Stable Diffusion, SDXL 1.0, for more realistic creations in films, music, and videos.
Now, developers can create agents for Amazon Bedrock to enable the automation of complex tasks. For instance, they can automatically access company data, like inventory, and convert it into a machine-readable format, helping service customer requests more efficiently.
AWS is offering vector engine support for Amazon OpenSearch Serverless. Developers will have an easy, scalable, and high-performing solution to build ML-augmented search experiences and generative AI applications.
They’re also introducing generative BI in Amazon QuickSight, enabling insights based on natural language questions. AWS HealthScribe will use generative AI to ease the paperwork burden for healthcare professionals. AWS is also making NVIDIA’s H100 GPUs generally available in production, ideal for running the increasingly complex LLMs.
According to the World Economic Forum, more than 75% of organizations plan to adopt big data, cloud computing, and AI in the next five years. To tackle this, AWS has rolled out on-demand skills trainings for those who want to understand, implement, and use generative AI.
