Science-fiction legend Isaac Asimov developed the now famous Three Laws of Robotics over 70 years ago. Since then they have been a guide for researchers in artificial intelligence.
Under them, a robot “may not injure a human being or through inaction, allow a human being to come to harm; must obey the orders given it by human beings, except where such orders would conflict with the first law; and protect its own existence as long as such protection does not conflict with the first or second laws.”
Just recently, the Financial Industry Regulatory Authority published its own three laws of robo-advisers in its Report on Digital Investment-Advice Report.
To call them laws, or even rules, would be an overstatement. The regulator even avoids using the term “best practices” in favor of “efficient practices,” which smacks of faint praise.
As long as digital investment-advice tool can understand customers’ needs, has a sound methodological grounding, and users can understand the tool’s limitation, Finra seems fine with them.
And it’s all the better if financial institutions document their internal processes and controls for these tools.
It’s doubtful that Finra, or the U.S. Securities and Exchange Commission, will ever go beyond principles-based regulation regarding investment-advice algorithms or their trade-execution counterparts.
Laying aside all the entire intellectual-property concerns, neither regulator has enough resources or internal talent to take a deep dive into the internal working of these algorithms.
For all intents and purposes,they remain ‘black boxes’ that only their architects and programmers truly know how they function.
Without this knowledge, the only way to examine the algorithms is to enter a statistically significant amount of inputs and examine the correlation of corresponding outputs.
This is probably why Finra chose to go the principle-based route. It’s much easier and more familiar to quiz members about their processes and controls than to look at millions of lines of code under a microscope.
Unless a rogue algo is responsible for triggering a market crash like 1929’s Black Thursday, regulators likely will keep prescriptive regulation off the table for now.
Featured image by Tatiana Shepeleva/Dollar Photo Club
