Twenty years ago, your family probably would have laughed you out of the room if you predicted that we’d have one day have internet-enabled lighting. Lo-and-behold, smart lights are here to stay.

These advances are evident everywhere in our lives, from smart cars to impending smart cities. But such technologies aren’t without their problems. Read on for three challenges we face stepping into our technological future.

1. Driverless Cars

At their most basic level, driverless cars sound like a simple idea. Take a car, pack it with sensors and give it a specialized computer for navigation, and you’re good to go. Right?

In reality, it’s much more complicated than that. As companies release more and more driverless prototypes, they’re also wrestling with mounting legal, technical and ethical issues.

For example, future driverless cars will require some kind of internet connectivity to communicate with each other and make quick decisions. They’ll also need detailed maps to drive accurately. Ethically, there’s the question of who shares the responsibility for an accident.

Another issue these smart cars face is what kind of fuel they will be built to run on. Some say they will almost certainly all run on electricity, which currently only applies to 14 percent or less of cars on the roads today. While the switch would certainly be a positive move for environmental reasons, some consumers will be likely to push back.

2. Artificial Intelligence

The media and popular culture alike have made the idea of AI, artificial intelligence, pervasive. Tech companies tout the advances made with their personal assistant software, and movies portray robots with personalities indistinguishable from humans.

However, we’re far from developing anything close to true AI. The amount of data needed to “teach” an AI is far larger than what humans need. Researchers can acquire data for image recognition easily enough, but data from other areas — like health care — is hard to come by due to restrictions and regulations.

Another issue is that AIs can really only be efficient at one task. For example, researchers can teach an AI to play one specific game, but they will need to retrain it to play another. What’s worse? To play the first game again, the AI would have to relearn it.

At present, we haven’t figured out how to make an AI that can perform two unrelated tasks. This, and the amount of data required, are major barriers to our progress. While we may be able to develop such systems in the future, we have a long way to go before we reach the digital assistants of our dreams.

3. Virtual Reality

In the game development world, VR, virtual reality, is the next big thing. Specialized VR headsets are all the rage, but VR still has some growing to do before it really comes into its own.

At the moment, less than one percent of the world’s computers have the processing power for virtual reality. While consumers have the option of building their own rigs, the project isn’t for the faint of heart. Specialized parts are expensive, even before the cost of the virtual reality headset itself.

Another relatively unexplored issue is potential health effects. Due to its immersive nature, virtual reality sometimes induces headaches and queasiness. While these symptoms are temporary, there haven’t been any long-term studies on virtual reality. While we know short-term use is fine, extended playtime could lead to unforeseen issues.

While the future is here, keep in mind we’re not yet at a gleaming utopia. All new technology has its drawbacks, so it pays to approach the latest and greatest with a grain of salt.

The opinions expressed here by Inc.com columnists are their own, not those of Inc.com.