As AI becomes more ubiquitous in the market, Qraft Technologies is using it to select stocks and balance its ETFs. Weldon Rice, Head of AI ETFs at Qraft Technologies, appears on Catalysts to discuss how Qraft is using AI and how AI-enhanced ETFs are different from traditional ETFs.
Rice emphasizes that artificial intelligence is used purely to construct the portfolio. However, he points out that the models themselves are built by humans: “We have experts in-house who help us decide what data points are being created, what types of models are being used, and drive that as we do more research.” Once the AI models are created, they use new data each month to make investment decisions, he explains.
Rice points to a recent breakthrough from Qraft's partnership with LGI Research, where AI models can now create reports that explain to investors why they picked certain stocks. “We see this as the next step in AI innovation in AI-driven investment decisions,” he explains. He adds, “The way artificial intelligence looks at the market is completely different than a human would. We're just looking at the market purely from a data perspective.” These models provide a nonlinear view of the market that humans can't normally see, giving Qraft's ETFs a unique advantage, Rice explains.
To learn more about expert insights and the latest market trends, click here to watch this full episode of Catalysts.
This post was written by Melanie Leal
Video Transcript
So, I'm curious, since you mentioned this is an AI-driven ETF, does it strictly rely on AI technology to make these types of decisions?
To what extent is there a human element, human involvement?
And is there any basis for some of the moves that have been made with your money in recent weeks?
absolutely.
So, you're right about how to build a portfolio.
We are using purely artificial intelligence.
So clearly, right now humans are in the model building stage.
So, you know, we have in-house experts who are helping drive that as we go through our research, what data points are being created, what types of models are being used, that sort of thing.
But once the model is created and trained, we never change it because the model gets new data every month and makes decisions based on its training.
So yeah, when it went through NVIDIA, if I fell that much, there was no one to say, oh, that's the wrong direction.
Uh, we need to change that in terms of the rationale behind it. Uh, actually, one of the big challenges with artificial intelligence is that it's somewhat of a black box because we're using deep learning models that look at very complex relationships, relationships in the data.
So that makes it a little bit harder to understand.
However, one breakthrough we have seen recently is not yet in the field of QR FT.
Um, but our most recent fund that we launched in November with LG I Research has an element of using a large language model, basically like CE PT.
And now LG I Research can actually produce a report that looks at why the AI model picked the stocks it did.
So, it's actually quite interesting.
Well, actually, we've just started posting that on our website now for investors to see.
And we see this as the next step in AI innovation towards AI-driven investment decisions.
What is the math that AI uses that investors aren't using to research NVIDIA?
So I think the way artificial intelligence looks at the market is very different than humans.
First, we look at it purely from a data perspective.
But, you know, if you look at the way that has typically been done in the past — either fundamental managers or traditional quantitative managers — the models have typically been fairly linear.
What artificial intelligence can do is, again, look at the relationships, look at the data, but also look at the relationships to the data.
And it gives you a non-linear view of the market that humans can't see.
And I think that's what makes it so unique.
And I think the real promise for investors is that they can get a completely different view of the market in their portfolio than they would get from a traditional human active manager.
Now, Waldon Rice Craft Technologies, the head of the AI ETF, is very interesting.
We are very grateful for your participation.
