Sensors are nearly everywhere, from our fitness bands and cars to dishwashers and thermostats, and from smart home automation to machines with "feelings" that can tell you when a part needs replacing.
Tapping into all that data has proved profitable in a number of previously unimaginable ways. Farmers have optimized their efficiency by knowing more accurately the state of their crops and when is the best time for watering and harvesting. Transport systems put out just the right number of buses and trains to conserve fuel. The list goes on.
But what about investors? Little has been said of incorporating all the Internet of Things sensor data into investment strategies, probably because few are trying and even fewer have found success. (If they have, they intend to keep their secret sauce guarded.)
But imagine the possibilities. Dr John Bates, group executive board member at Software AG, had a few potential scenarios to share. For example, on massive farms, there are tractors that drive themselves and have sensors to convey the state of the harvest. "If you can tap into all that data, maybe you don't need to wait for the crop reports. You can do your own automatically with analytics." But it poses a big question: "Is this a new way of insider trading?"
And it's not just tapping into trade data. Machine learning can also be used to spot when unusual things happen. Envision, from an HR perspective, combining more traditional data like holiday requests and travel schedules with sensors that track where traders are in the building, what time they've arrived, who is meeting with them, and how often. And now there's smart video analytics that can tell facial expressions like sadness, worry, or happiness.
"What if they could correlate that stuff?" Bates muses. "If an unusual trade happens just before news article came out that moves markets by a certain amount, and the trader had been meeting somebody outside the second floor elevators looking nervous, and this has happened three times in the last week, we investigate. This is going to lead to massive riots... Traders will say this going too far."
Regulators leveraging IoT
Understanding what the IoT could mean for traders is important for regulators to get ahead and define some of the legal grey areas that might emerge.
Using the examples above, some legal gray areas might be whether it would be okay if a third party aggregated the farm information, combined it with satellite imagery of fields, and sold subscriptions to trading shops? If so, would that service come under regulatory scrutiny? And if that data could affect share price, how public would this data be?
Would regulators call foul on firms that could not correlate the sensor data and flag suspicious employee behavior? What responsibility would a firm have to adopt these surveillance measures?
"Regulators are going to be tapping into all these techniques and speeding up," Bates said. It is very probable they will leverage sensor data to track more people and things, just as firms will use the data to innovate their strategies.
Sensors for protection
Bates offered another hopeful prediction: Could algorithm-generated problems soon be things of the past?
Many flaws in algorithms have been revealed (very publicly), and a determination to avoid attention from the news media and regulators has meant a lot of preventable problems have been attended to tirelessly.
Today, smarter self-learning algorithms are spotting instances when things are deviating from normal patterns. In reaction, they know to shut down the instances, report problems, and, most importantly, not make drastic and hasty decisions, such as ones that can cause flash crashes in the market.
With all the safeties in place, "I think we're being so boring now," Bates said.Becca Lipman is Senior Editor for Wall Street & Technology. She writes in-depth news articles with a focus on big data and compliance in the capital markets. She regularly meets with information technology leaders and innovators and writes about cloud computing, datacenters, ... View Full Bio