Data & Business Intelligence

Data & Business Intelligence
What advice would you give to enterprises regarding data strategy?

Top Answer : At the enterprise level, data is another piece that will benefit from startup processes and policies. These need to be implemented as data becomes a large issue if it’s not managed efficiently. In general if there is 100 GB of total data in an organization, 30 GB of that is not required at all. It’s duplicate data that’s been copied several times in several places, often because somebody copied it for a backup or some other purpose. Retention data are stored for a much longer period and in a much more efficient disk, which is beyond regulatory or organizational requirements. Backups are generally taken multiple times in different formats, which creates a lot of confusion in terms of restoration or modernization work.  A lot of applications, data and configurations within the huge data center platforms are never even used. It's just in the system storage; nobody’s used it or bothered to clean it. A lot of people don't know this kind of data is there. There has to be some data cleaning and application processes, as well as strict measures to follow and automate most of those processes. There needs to be a team to sort out what is required or not; then the system will have much less data. That means we'll use the machines faster, applications will be faster, user experience will be very good and NPS will be high. Today, people cannot think innovatively in a large organization because they know that if they want to implement something, there’s a lot of work to put in. And this thinking creates a blockage: "I cannot work on this, there is no value. Why bother thinking about this?" Then nobody implements anything new.  All the large enterprise companies need to adopt the startup-style approach to workforce and architecture—organizational architecture, as well as system and data architecture. From one organization to another there are different processes but if a large organization can do these things, they will benefit in terms of their employees, costs, customer satisfaction and digital transformation.

ERP Usage and ChallengesERP Usage and Challenges

How are decision-makers using Enterprise Resource Planning (ERP) tools? Benchmark with your peers.

Does strict data compliance mean compromising the speed of business transformation?

Top Answer : If you keep your system, applications and data very thin and lean, then you will have less data, fewer applications and fewer people. That actually makes security implementation very easy. If we go from having 100 machines to having 70 machines or even just 13 machines, then security implementation is easy and you have less data to protect. The data leak chances are lower than if you secure just 1% of the data when that same data is copied to another hard disk, which was left open after somebody copied it for testing or some migration. That data is still laying there open because the job was to copy it, not to delete it from the old device. GDPR and other data retention policies are clear about how long you have to retain the data and all industries are more or less standard so there is no ambiguity. But right now nobody deletes the old data. For years and years after the required period, all that data is stored in their system. Maybe it is stored in two or three locations—that is where it gets complicated. If you have less data and applications, then compliance is very clearly defined and we all know what to do.

What are your biggest concerns as AI capabilities become a standard industry offering? 

Top Answer : If AI is done right, it's a huge opportunity for technology; if we get it wrong it’s a huge slippery slope. There's already been evidence, in some cases, of the consequences involved if we mess up broad aspects of AI: a few years ago, Microsoft put out a chatbot that didn't have the right monitoring and people were seeding it with harassment and hostile language. That basically trained the chatbot to say inappropriate things back to people because people had flooded it with all this negative stuff. There have been instances of AI that have already had severe consequences, like the criminal risk assessment algorithms that are used by some courts to determine if people will reoffend. There was a research study done by a group of PhD students that found that this system was substantially discriminatory against African-Americans because of the way in which the data was put into it. It used machine learning to determine somebody's likelihood to recommit crimes based upon the anchor point of a historical bias, which the machine then learned to continue going forward. I've worried a ton about that at Cylance and even during my time at Intel.

Have safety concerns or security risks stopped you from implementing AI technology?

Top Answer : In the early days of WiFi, when we were just transitioning to it from hard line, I used to tell folks that if it’s properly configured, implemented, and locked down, WiFi is as secure, if not more so, than a lot of wired systems. But in the early 2000s, I didn’t encounter a lot of WiFi that was properly configured, implemented, and locked down. I would copy/paste that analogy here to AI.  If it is properly implemented with proper principles and properly managed and maintained, then I'll take AI wherever I can get it. I don't have enough bodies, resources, or time to manage the multiple petabytes of data that come my way and I want to be able to refocus my teams to do it. But all of those caveats need to be in place. My challenge isn't where it can be useful, it’s that I want those principles in place and don’t see a lot of them right now.

What strategies should IT leaders use when they engage AI vendors to minimize risk potential?

Top Answer : On the security side we know we can benefit from it and in some cases we already have proof of value. But we have to do enough as purchasers of AI-based security products to ask the security vendor questions around their principles and practices, such as, “Do you have all the data? Can you get all the data?” Let's say I'm doing data protection and want AI capability. Is it okay to use data on the internet that may have been stolen and leaked in order to train your model? That’s a solid question to ask because if you're training your model on other people's sensitive data that was leaked, it was never intended to be out there. I know some organizations that have almost gone down those slippery paths; there are probably a few that have but that’s not happening on my watch. It's an ethical choice but asking where they got their data from is very relevant if we’re buying an AI-based or ML-based data protection capability. It would give you an idea of an organization's moral compass and whether or not they're pointing in a direction that's consistent with yours.

7 views
2 comments
1 upvotes
Related Tags
Is open source a cybersecurity minefield for AI?

Top Answer : The open source tools are great but you have to ask, how have they deployed it? Where are they getting the data and where are they hosting it? A few years back I did some work for a company in the AI space which was basically an amalgamation of open source tools hosted in your own dedicated Amazon VPC to be able to run whatever you wanted. In that case it was pretty straightforward but in many instances of AI—especially with AI providers—that data is probably coming from your networks or systems and it’s going somewhere else, probably to a cloud service that will have the compute power. If they are actually doing ML, you need some good compute power and the cloud services are great for that.  But chances are it's not one cloud: they've got some compute power over here doing this analysis, then they're sending the data over to a Tableau instance, which is sitting in another area. Having the tools is great because it gives researchers the ability to do a lot of things and to improve the tools. But it's a question of where they’re deployed: How can you get your arms around where the data is residing as well as the data about the data? Machine learning is taking data, doing all this analysis, and creating more data, so where's that data sitting?

Does machine learning (ML) use artificial intelligence (AI) or does AI use ML?

Top Answer : I see ML as something that's actually looking at my dataset and my environment, understanding what's going on and then surfacing that data in a way that is helpful—maybe in a way I didn't know I needed that data. The AI is the conversational part that is acting on my behalf, executing whatever I want it to execute. That's where I make that separation. I'm much more excited about the ML data side of things than I am about AI at this point.

Digital Employee Experience During Remote WorkDigital Employee Experience During Remote Work

With a highly distributed workforce, employee engagement is crucial—and digital employee experience platforms are helping Pulse community IT leaders prioritize this.