Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The Internet had gone mainstream by the late ’90s, but Miro Mitev was headlong exploring something that wouldn’t become popular for decades: AI.
Now an asset manager, Mitev was an early adopter of AI in finance after discovering the capabilities of neural networks in 1997 while studying at the Vienna University of Economics and Business.
He told CNBC that he saw the potential of neural networks for financial forecasting. “I fell in love with these kinds of possibilities,” he said.
Mitev spent his 25-year career doing forecasting for banks and technology companies like Siemens. He founded SmartWealth Asset Management, whose decisions are made entirely by a network of AI systems. Its latest fund, IVAC, targets $2 billion in assets under management and has an annualized return target of 14% to 15%.
Despite the lack of human involvement in AI decisions, Mitev said “humans are the most important part of the equation” because they are the ones who select the training data, input the variables, build the parameters and constantly tweak the model.
Once a pattern is created, “it’s very dangerous to start intervening,” Mitev said. Indeed, trusting the model is his golden rule, he added.
Instead, humans should ensure that there are no errors in the data or calculations and introduce new data so that the model is up to date.
“The worst thing is to cancel the results, and that’s what happens very often,” Mitev said, adding that at first people “don’t trust” AI. “Even though we as humans don’t see the result now, if we look back after two or three months, we say, ‘Oh, actually we were wrong,'” he added.
The forces that drive the market – optimism, pessimism, speculation – are very human. Even the European Central Bank has warned that the current rise in A.I. may be motivated not by detailed technical analysis but by the fear of missing out.
Mitev said that removing emotion from investing shows better results; SmartWealth Asset Management recorded gains of 407.63% over a 10-year period through November 1, 2025, compared to an industry benchmark of 145.34% over the same period, according to a chart shared by a company representative with CNBC.
It’s “not possible” to know what will happen a year from now, Mitev said, but he can see up to a month in advance with his model. “Evaluating this information and making informed decisions based on it is consistently shown to provide better results than humans.”
Constant monitoring and the introduction of new data are important points, given that AI systems “hallucinate”: generate false information. Mitev said model errors were due to “overfitting,” data issues or model misspecification.
Overfitting is where the algorithm pays too much attention to what Mitev calls “noise.” He said it was data “that doesn’t make sense” because it doesn’t reveal a true cause-and-effect relationship with stock performance.
Rigorous design, validation and real-world testing serve as an antidote to this, Mitev added. This means that even though its fund strategy is executed entirely by a series of algorithms, humans still play a crucial role in ensuring its effectiveness.
“It’s actually a process that evolves over the years… and that’s why internal development of these types of technologies is very important,” he added – especially for anyone looking to differentiate their game when it comes to AI.
