AWS re:Invent: AI baby steps are cool (and crucial for success)
You’ve heard a lot about AI this and AI that over the past year or two. I get it.
But where are we now? And where are we going?
At AWS re:Invent in December, I led a session alongside Phil Taylor , our CTO of Digital Velocity, discussing today’s AI market — from Amazon Q to NVIDIA and everything in between. Let’s just say it’s an incredibly awesome time!
Start with a strong foundation
In 2023, the global AI market was worth around $196 billion. Grand View Research estimates that’s going to jump 36% by 2030.
The beauty of AI and generative AI is that they can help you solve some very real problems. You just need to know what you want to achieve first.
As Phil and I discussed on stage, there are multiple foundational models that can help get you where you want to go without an involved training cycle (think AWS solutions like Bedrock and models it hosts like Meta’s Llama, Anthropic’s Claude and AI21 Lab’s Jamba). As we’ve collectively thrown things against the wall as an industry, a couple of use cases have stood out so far.
1. Drive internal efficiencies
Here at CDW, our sales teams are tasked with becoming experts on hundreds of products. And we’ve found a way to augment that vast knowledge with the right intricacies for each customer’s needs. (SPOILER ALERT: It’s AI!)
To make it happen, we combined public-facing documentation with our own internal curation, put it into a vector database and hooked it up to a chatbot using retrieval augmented generation (RAG) to improve query results that come back from those Bedrock-hosted models with far more focused answers.
Plus, by utilizing AI internally, you have far less to worry about when it comes to leaking any sensitive data out into the world inadvertently.
2. Take it to the edge
Another exciting use case for AI is in edge computing. In fact, ResearchAndMarkets.com predicts that the global edge AI market is expected to reach a market size of $59.9 billion in 2029 from $23.5 billion last year.
Recommended by LinkedIn
We’ve all had the experience of walking into a fast casual restaurant, ordering from a kiosk and then seeing your bag pop up ready to go on the counter, right?
With edge AI, we’re going to start seeing instances where when you walk through the door, it knows who you are and what your last orders were to jump-start the ordering process. And companies like NVIDIA, Intel and even Apple with their M4 chip are supporting edge hardware for these foundational models.
When you combine edge hardware with the foundational models and RAG, you can implement AI anywhere. We’re going to start seeing it cross over into all phases of our lives soon… not in a Tom-Cruise-in-Minority-Report way but more in a this-is-actually-very-useful-and-not-scary way.
3. Identify your biggest challenges first
More than any other technology I’ve seen in my 30+ year career, there’s one clear way to gaining AI buy-in and moving ahead with confidence.
Begin by outlining the 10 biggest business problems you’re currently facing with an emphasis on those that improve internal productivity. You don’t need to solve all of them at once. Instead, look at your data next. Out of that initial 10, you likely have the right data quality to solve three of them. Pilot one of those as an AI learning experience. Rinse. Repeat.
And with CDW on your side, we can find opportunities where AI can shine across your organization.
What’s next…
If you don’t have clean data in place, AI won’t be able to solve your business challenges. Think of it this way: AI is still a garbage-in, garbage-out world. You can identify every problem under the sun, but if you don’t have the right data lined up, you’re out of luck.
After we returned from AWS re:Invent, Phil and I sat down to discuss the biggest takeaways and highlights from our session. Check it out here.
Viva Las Vegas!
Later this year, we’re heading back to Vegas for AWS re:Invent 2025, and I’m excited to see how AI has continued to develop and grow by then.
Got questions about how CDW can help you take the next step in your AI journey? Or maybe your first step? We’re here to help. Reach out anytime and visit CDW.com/AI for even more.