Artificial Intelligence (AI), broadly defined as computers doing the work (and perhaps even reasoning and thinking as) humans do, is having a moment. Well, it’s been having a moment. Industries from retail to entertainment to automobile manufacturers are experimenting with intelligent machines and the technology is set to become more and more a part of our everyday lives.
Even Mark Cuban, famed businessman and entrepreneur (and The Zebra investor) said last week that AI is the future of content and entertainment, and that “the world’s first trillionaire is the person who masters AI and all its derivatives.”
What is AI, Anyway?
The term AI is thrown around a lot, especially lately, and is often used in conjunction with machine learning (ML), however there’s an important distinction (from Forbes): artificial intelligence is machines carrying out tasks in a “smart” way while machine learning is an application of AI in which machines use data to learn for themselves. Developers of AI technology aim for machines that can think and even reason just as humans do.
AI can be broken down into two types: Generalized AI and Applied AI. Applied AI is more common and involves systems that can intelligently carry out specific tasks, like trading stocks, says Forbes, adding that autonomous vehicles* would fall into the applied AI category. Systems and devices that could handle any task would be defined as Generalized AI, or more simply, “the current state-of-the-art” AI.
Still, systems and programs using AI aren’t alive or conscious in the sense that people are. Writes TechCrunch, AI systems “have no genuine creativity, emotions or desires other than what we program into them or they detect from the environment.” And, unlike AI systems in the movies, real artificial intelligence-powered machines would have no innate desire for families or mates (or to take over the world). Think of AI programs as highly trained specialists with no inborn ambition of their own – specialists humans still ultimately control.
The Latest AI Developments
Truly exciting uses for artificial intelligence have already entered the market, and as both AI and machine learning mature, we expect cool innovations to abound. A few of our favorite:
- Computer scientists in Canada developed an AI-powered system called DeepStack that can beat top professional Poker players. The system not only reasons but uses “intuition” developed through deep learning and repeated strategy assessment.
- Facebook now uses AI to identify users at risk of suicide and (after a team of actual humans gives the go-ahead) offers ways the at-risk person can get help.
- Domino’s pizza has adapted AI in its effort to boost sales of their pies. The AI-powered systems will help customers find coupons and order from literally anywhere by dropping a pin. It also helps stores restock based on anticipated demand.
- AI is even helping people with tortuously slow internet connections view videos that actually look good. Netflix is introducing AI techniques that will make the pauses and other video disturbances common with slow connections a thing of the past. Look for it in the next couple of months, company insiders say.
AI and the Auto Insurance Industry
The potential for AI in the automobile industry is extraordinary, as is already being demonstrated by major advancements in driverless car technology. From ReadWrite: “Just like a human, self-driving cars need to have sensors to understand the world around them and a brain that collects, processes and chooses specific actions based on information gathered.” AI comes into effect for such critical functions as refueling or recharging a car when it needs it, adjusting directions as new traffic data becomes available, and using speech recognition and eye movement for enhanced communication and safety.
But how does AI impact auto insurance? Insurance companies appear to be making their first AI moves on the collision and claims front.
In January, news came from Solaria Labs, which supports Liberty Mutual Insurance’s technological developments, about its new AI portal which aggregates public data to “help users find potentially safer routes — and help them assess damage when that doesn’t work out,” reports TechCrunch.
Solaria Labs is developing an “AI Auto Damage Estimator,” which would help users determine on the spot after a collision which repairs are likely to be needed and how much they’re likely to cost. The AI in the Estimator would compare damage to a user’s vehicle to that of thousands of anonymous claims photos to provide the estimate.
TechCrunch writes that the API would also aggregate a wide variety of public data for auto theft, parking citations, and crashes to help drivers find safer routes and parking places without the normal hassle.
Will AI Bring Auto Insurance Savings?
As for the relationship between AI and the rest of the auto insurance industry, the jury’s still out, but The Zebra’s licensed auto insurance agent Neil Richardson explains that AI has the potential to save auto insurance customers money. Specifically, if auto insurance companies can save money by turning claims investigations, insurance adjustments, and repairs over to AI-powered machines, they’ll be able to operate at a lower cost–savings they could pass on to customers.
“The cost of doing business for each insurance company is factored into the rates that they charge their customers,” Neil explained. “Anything that can bring that cost down, should make a huge difference in the premiums companies can offer consumers.”