Betting on AI and Blockchain? Some notes here
Dimension Data’s APAC CTO, Andy Cocks, has opined that sooner or later, people will begin to notice blockchain as a technology that is potentially transformational for other business models, besides just cryptocurrencies. He said, “The underlying use cases for blockchain, are quite interesting.”
He was sharing about Dimension Data’s advocacy of the use of blockchain technology, for security in Internet of Things (IoT) applications.
“It all starts with IoT. The sensors now are cheap, a few cents a-piece.” This is causing the use of sensors to become very prevalent… but how are they being protected?
Cocks also shared, “We saw (a network of) video cams being attacked last year, in Singapore, and a telco was taken down.
“You can’t run a full security stack on devices like these, so you typically run everything back to the data centre, which is a bit primitive.
“In this distributed IoT world, that is not going to work. We believe, the use of blockchain-type technology with security is going to be pushed out, because blockchain can provide inherent security and support micro transactions.”
According to Cocks, blockchain technology is cost-effective enough to charge 0.01 cents for example, to provide security with technologies that are pushed out to the edge of the network, like video cameras.
“What we are advocating is, the security industry will start to embrace blockchain-like technologies, to provide security for devices at the edge, as opposed to in the data centre. That’s our CTO’s hunch,” Cocks said, adding his opinion that IoT won’t take off till its more secure, but that’s currently an expensive endeavour.
“Blockchain is the answer to this.”
Artificial intelligence – early days
The APAC CTO also had a fair bit to say about artificial intelligence. For one, everyone is dipping their toes into it, but it’s still very early days, and in the proof-of-concepts pilot phase.
“But, you also have to ask yourselves, what is artificial intelligence (AI)?”
In the new digital economy, this question is still being answered, and a good place to start would be the tools that come from AI like speech recognition, machine learning, image analysis virtual agents, and many more.
Cocks opined, “IT owes it to the business units, to help educate them as they want they want to do with (AI tools).
“Just AI, is not good enough. IT’s role for example, is to put a platform-as-a-service (PaaS) in place for users to tinker with all these tools.”
But underpinning all of these AI tools, is data. “You can only start to get intelligence, if you have all of the data.
“The biggest challenge businesses have when implementing AI, is to consolidate all of the data sources in the organisation, into data lakes.
The issue arises because most systems that generate these data, are in siloes or stove pipes.
“So, an organisation’s initiative to build a data lake, and make it available for business users in an organisation, is still a big challenge.
To top it all off, organisations are supposed to communicate wot business units about what can be done with AI. “I don’t think they have even started to do that,” Cocks concluded.