AFME: Is AI harder to police than humans?
01 October 2018 London
Image: Shutterstock
A discussion about whether artificial technology (AI) is harder to police than humans split the Association for Financial Markets in Europe (AFME) Legal and Compliance conference audience right down the middle, with 50 percent saying “yes” and “no”.
The question was asked in a panel named ‘Overseeing AI and algo-trading’.
The crux of the conversation leaned toward taking machine learning away from the pilot space and more into everyday utilisation while making sure the next generation knows, and has access to, the tools to police it.
One panellist said: “The debate is about transparency, black box deception, all the way to deployment to ensure good governance around AI.”
The panellist discussed that this level of governance is, unfortunately, really only known by a “very niche number of individuals” who actually know what is going on.
They added: “A lot of people [in financial services] feel threatened by AI, but it is important to embrace it and look at business models. Really carve out where humans are needed. And then question at what point does it make sense [in your business model] to have a human there? And where does it make sense to use machines?”
The moderator also asked the audience if they thought regulators had moved quickly enough to address the need to supervise AI. The majority, 73 percent, said “no”, while 27 percent said “yes”.
In slight defence of regulators, one panellist said: “[Recently] there has been a lot more expertise in AI with regulators but it’s moving so fast.” She questioned: “How does anyone keep up?”
Another panellist discussed legal and regulatory frameworks lag behind in other industries, not just asset management. He used Uber and the music industry as examples where legal and regulatory frameworks have needed tweaking in recent years. He added this is a global issue because it is the “nature of the technology we’re dealing with”.
Another panellist concluded: “Technology is moving so fast, having a good use of technology data can give you an effective edge.”
In a speech after this discussion, a representative from the International Organization of Securities Commissions also reiterated the concerns of AI and the issues around machine learning changing at such a fast speed.
He questioned should we be putting in roadblocks or intervention, concerning AI? And in what places within the financial industry?
In a concluding point, he said: “I’ve probably left you today with more questions than answers, but it’s something we need to go away and think about.”
The question was asked in a panel named ‘Overseeing AI and algo-trading’.
The crux of the conversation leaned toward taking machine learning away from the pilot space and more into everyday utilisation while making sure the next generation knows, and has access to, the tools to police it.
One panellist said: “The debate is about transparency, black box deception, all the way to deployment to ensure good governance around AI.”
The panellist discussed that this level of governance is, unfortunately, really only known by a “very niche number of individuals” who actually know what is going on.
They added: “A lot of people [in financial services] feel threatened by AI, but it is important to embrace it and look at business models. Really carve out where humans are needed. And then question at what point does it make sense [in your business model] to have a human there? And where does it make sense to use machines?”
The moderator also asked the audience if they thought regulators had moved quickly enough to address the need to supervise AI. The majority, 73 percent, said “no”, while 27 percent said “yes”.
In slight defence of regulators, one panellist said: “[Recently] there has been a lot more expertise in AI with regulators but it’s moving so fast.” She questioned: “How does anyone keep up?”
Another panellist discussed legal and regulatory frameworks lag behind in other industries, not just asset management. He used Uber and the music industry as examples where legal and regulatory frameworks have needed tweaking in recent years. He added this is a global issue because it is the “nature of the technology we’re dealing with”.
Another panellist concluded: “Technology is moving so fast, having a good use of technology data can give you an effective edge.”
In a speech after this discussion, a representative from the International Organization of Securities Commissions also reiterated the concerns of AI and the issues around machine learning changing at such a fast speed.
He questioned should we be putting in roadblocks or intervention, concerning AI? And in what places within the financial industry?
In a concluding point, he said: “I’ve probably left you today with more questions than answers, but it’s something we need to go away and think about.”
NO FEE, NO RISK
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
