5 auto industry AI trends for 2021 and beyond

Driver: "OK Volkswagen, get me 10 gallons of unleaded. And a bag of fuego Takis please."
Car of the future: "Sure thing. Fill it up. Your Takis are on the way."

Despite the shortfalls of voice recognition technology, our robotic friends Siri, Alexa, and Google have made huge strides in getting things right. They've overcome privacy concerns, added digital voiceprints, and become multilingual. Big Tech's voice assistants are also learning new tasks daily through open-source APIs.

Intelligent voice penetration in homes is through the roof. More than half the U.S. population are millennials or younger and their preferences drive tech adoption. As Big Tech makes strides in digital voiceprint security, consumers will further adopt AI into more areas of their lives — including the car.

Though the digital commute has replaced the physical commute for many workers in the COVID-19 era, cars remain the primary mode of transport for most. But drivers and auto manufacturers need to move away from smartphones. One in four auto accidents each year — 1.6 million total — are caused by phone use. Integrated AI is emerging as the answer, and manufacturers are investing handsomely to make the technologies seamless.

Adoption of wireless connectivity for all kinds of apps in cars is over 90 percent. AI and social media recommendations inside vehicles are natural extensions of these capabilities. Would you appreciate it if your car recommended the lowest-priced gas stations when your tank starts to get low, or a place to park safely and cheaply when you're in a new city? Auto manufacturers think you might and are prioritizing these use cases for potential value-added features in their next generation of cars.

The market opportunity for connected car commerce is estimated to be over $230 billion. Recommendations for gas, food or parking is just the first step. Helping you pay for it without having to fiddle around with buttons is next.

Below is a summary of five AI trends you're most likely to see:

Research into payment-by-vehicle has already begun. But, to scale it, auto manufacturers will need to earn their customers' trust by demonstrating best practice privacy and data security measures. Manufacturers will also need to determine how consumers make that payment — RFD, Bluetooth or unique auto fingerprints.

Expect to see retail brands, including software startups such as Instacart and Shopify, dive into automotive partnerships. Imagine next-level curbside pickup, where upon reaching a store your car talks to their app to tell them you're ready. Or better yet, your GPS alerts the store when you are close, so your order is ready upon arrival.

Most car makers have integrated AI into the cockpit. General Motors' Onstar, for instance, can send commands to electronically disable a car that has been stolen. But, for most brands, creating their own voice agents may prove too ambitious. Seamless integration from castle to car with voice assistants and keeping the same platform solutions might prove more efficient.

On the flip side, there are a few manufacturers who have chosen to go their own way and invest in proprietary AI technology, like Tesla. Could you picture yourself one day saying, "OK Cybertruck, find me a roller disco"?

Improving the responsiveness, intuition and comprehension of AI assistants may be the biggest demand for most consumers. However, relevant advances are on the way may help in:

• Remembering established preferences and patterns
• Accurately understanding the context of questions
• Answering questions more quickly

The speed issue may be solved with ubiquitous 5G, which is expected to roll out over the next few years. In fact, some believe the automotive industry represents the largest market opportunity for 5G solutions.

Having your car hacked or stolen is a major concern for more comprehensive AI systems. What if someone records your voice and simply replays it in order to get in? With the advances of deep neural networks, having just a sound bite might be enough.

In-car AI systems must be both through and rapid in recognizing and identifying you. Further improvements in fingerprint, retina scan, and voice recognition technologies are needed to provide smart and elegant solutions to the auto security.