There are no shortage of posts flooding the ecosystem about NVIDIA GTC. With 35,000 attendees, it’s hard not to take notice. GTC has become the must-attend event for AI enthusiasts and spans everything from chips to coding agents. I’ll be doing a full unpack of all the announcements and innovation, but I wanted to start by reminding folks that NVIDIA GTC, today, is not an Enterprise AI show.
The trick to understanding NVIDIA GTC is what the expectations need to be going in. There have been no Enterprise AI stories but there are also no Enterprise AI companies. I’ve been to a lot of Intel and AMD events and there is a marked difference in the content, sponsors, and audience.
GTC is like the early days of AWS re:Invent. We have a plethora of AI vendors in the ecosystem. They are looking for every opportunity to exploit a gap, and using raw excitement combined with a growing token bill to build and prototype as quickly as possible.
The Proliferation of Prototyping
Prototivity (noun) – /ˈprōdəˌ tivədē/ –
— Eric Wright (@discoposse) March 17, 2026
Generating unending amounts of vibe coded prototypes to optimize available tokens without actually focusing on bringing anything to market.
See also: Token Anxiety, Agentmaxxing, Tokenmaxxing
AI use-cases are everywhere. It’s different than previous waves of hyped innovation. Not too many years ago, a huge rally of builders were wrapping around blockchain. While blockchain as a technology is still important, it didn’t unlock the “future of work” in the way that AI and agentic AI are today.
OpenClaw (nee Clawdbot) became something that even normal non-tech folks were becoming aware of. The fastest rise to popularity has not slowed yet. The introduction of NemoClaw was a strong validation that this deployment and operational pattern is very real.
But if you think that your healthcare provider or financial services provider is racing to put this in their production IT portfolio today, the data and proof points are not on your side.
It’s Still Early for Enterprise AI
As amazing as these innovations are, they are also still at the peak of early adoption. We haven’t even begun the famous roller coaster ride towards the trough of disillusionment as Gartner coined in their description of adoption lifecycle.
That said, most of the enterprise analysts are battling for their own oxygen right now as AI potentially eats large chunks of their value proposition. When I read articles like “analyst predicts YY% of <thing> will be replaced by AI by 2028” and that same analyst was wrong last week about something this week, I have to wonder…well…I’ll leave that for another article.
The point I’m making is that what we are seeing today is not THE shift. It’s the beginning of the shift. Consumer adoption of AI technology, whether desired or not, is happening at a breakneck speed. Just about everything you do online is touched by AI in one or more ways.
Agentic AI is amazing, but any use in enterprise settings is still early. Security in agentic AI is a tire fire at best, and a global risk at worst. Foundation models are no longer in the weekly game of one-upping each other, now it’s about features on top. The combination of both those things also creates a huge opportunity for sovereign AI.
I’ve been working on helping sovereign AI companies for years already. This is not a new concept, but broad adoption of the pattern is. I’m excited about our opportunity ahead.
The real stories that will move the industry forward are going to be the ones where enterprises tell us how they are doing things, not vendors telling us how we should do things.
NVIDIA GTC is Fantastic, and Will Only Get Better
You have to give kudos to NVIDIA just on the sheer volume of activity and opportunity that they have created. The ecosystem is thriving (some would say a little too much), and there are amazing stories of very human-powered successes that we need to keep our focus on.
Next year NVIDIA won’t be able to just be “more NVIDIA”. Going forward, GTC will need proven enterprise stories, and more implementation partners with skin in the game. As fast as we are moving, it’s still early days.
I’m looking forward to hearing and seeing more from the event. Congratulations to Jensen and team on a fantastic event and also to the community of practitioners who are turning NVIDIA innovation into our opportunity.
