2017 was a year of AI. 2018 will probably be no different. Money is flowing into AI startups, often irrespective of whether there is a “real” AI involved. The AI hype wave is high, and companies are riding it, with success. To discuss this status quo, we had a chat with Peter Jidesten, experienced entrepreneur, co-founder and CEO of iDefendo, a Blockchain-based File hosting service that creates digital witnesses for files, shares etc.
– Where do you draw the line between the hype and actual value?
P: As is often the case with new and possibly disruptive technologies there are both risks and rewards for those who get in early. As a rule of thumb, it’s often better to focus more on how well a new technology solves a real and actual problem, rather than on the technology itself. If you see actual benefits from some tech, then it’s, by definition, not just hot air or hype.
– There’s been so much confusion between AI and automation. Does this usually concern investors?
P: I can’t speak for everybody of course, but the investors we’ve been talking to have really been interested in getting their heads around what can be accomplished. They have all been trying hard to really understand. That’s a good sign, because I don’t think anyone wants to go back to the feverish frenzy of yesteryear’s tech bubbles.
– In your view, how many AI startups focus on delivering actual value? Is the situation similar in the blockchain ecosystem?
P: Both the AI and the blockchain ecosystems have relatively low barriers to entry. That naturally leads to companies getting started without extensive analysis taking place beforehand, and to a certain ‘we’ll build it because we can’ attitude. But that’s not necessarily a bad thing, as it also leads to very rapid experimentation that very well may lead to actual values being produced, although not always in predictable ways. In fact, much of what we use and love on the Internet today was built in this way. But the ratio of failed to successful endeavors is probably disheartening to many.
– Where do you see the blend of AI and blockchain the most useful? Are you aware of any real-life examples?
P: As I mentioned earlier I think it’s important to look at what problem a technology solves instead of focusing on the tech itself. Blockchain solves the problem of trust between parties who don’t know each other, where there is low or nonexistent implicit trust. Money did that too, back when it was invented, and it’s probably no coincidence that blockchains came into the spotlight attached to virtual currencies like Bitcoin. It seems plausible to me that the trust-producing aspects of blockchains will be useful in AI solutions as well, facilitating trade and cooperation between entities that don’t necessarily have done business with each other previously and who might not ever meet in real life. Or who don’t even necessarily exist in the physical world.
– How to make clients see through the hype?
P: By teaching them to ask the right questions I think. Clients should focus more on how well a solution solves a problem than on the tech itself, just as they do when evaluating other kinds of machinery. AI and blockchains are by no means snake oil, but that doesn’t mean that there aren’t snake oil salesmen out there, busily selling the hype. Luckily, critical thinking gets you far!
– In your startup experience, how important is it to have a lawyer involved from the start?
P: For a startup, having a lawyer is sort of what having fire insurance is for a home owner. Sure, your house will probably not burn down, and your agreements and contracts will probably be just fine. The problem is that if you’re not that lucky, your problems quickly become catastrophic. That can easily be the end of your ventures, even if everything else is going your way. And there is also a good chance that having a lawyer on board from the very beginning will make you look better if you find yourself in a due diligence down the road. So I think it’s very important, and I think you need to find a lawyer who ‘gets you’ and who you can have a good long-term relationship with.
– In your opinion, do we need more legislation around AI? What about blockchain and its applications?
P: Right now I don’t think that anyone really knows enough to be able to legislate wisely. So what I think is needed now is for legislators to take an active part in the development of both AI and blockchain, so that their perspective is part of the efforts as early as possible. Because at some point legislation will happen, and I think that legislation will probably be better if it’s developed in tandem with the technology it’s supposed to regulate. AI has the potential to cause massive disruption, so we need to look at it from every perspective.