When it comes to emerging technologies, we get lots of questions as tech professionals, business leaders and boards of directors grapple with the ever-evolving nature of the business and how to best determine a future course for successful growth. During a recent webinar dealing with the impact of emerging technologies, we received a number of questions we didn’t have time to touch on during the Q&A session. We answer a few of them here.
Q. How do you recommend large companies deal with IoT accountability or ownership?
Scott: It’s important that organizations, regardless of size, establish both a governance committee and an approach to IoT that involves the business and IT. Just as with any other technology asset, IT should be considered the steward of the asset, not necessarily the owner of the asset. As the steward, IT drives the protection, maintenance and updating of the tool. Initially, the governance committee will define the roles, accountability and the approach that will be implemented. We often see breakdowns occur when areas of the business are purchasing IoT technology; IT needs to manage that technology but doesn’t have the accountability. While this isn’t limited to large companies, it can be exacerbated in bigger organizations, where it may be harder to get control of and easier for things to get lost. Same issue, just a larger scale.
Q) How is quantum computing applied — is it through cloud computing?
Scott: At this time, most organizations are adopting a hybrid approach, running certain functions on quantum hardware and certain functions on classic hardware or cloud. There aren’t a lot of quantum computers out there right now, so most are opting to rent time on existing quantum computers or through cloud-based environments, vs. building their own quantum setup.
Current quantum computers lack enough quantum bits, or qubits, to rival classical computers just yet. But one approach that promises to bring potential advantage in the near term is optimization. Experiments in portfolio optimization have caught the attention of leading financial services firms, which already are making significant investments in this space. For example, with such a hybrid approach in financial services, classical computers can parse tasks to a type of quantum computer called an annealer, which then analyzes data to identify profitable trades. The classical system then pieces the results together.
Beyond financial services, other optimization problems this hybrid approach can tackle include finding the best shipping routes, such as for delivery trucks.
Anywhere we use machine learning today, we can expect to see some quantum edge in the future. Financial services firms are also interested in improved fraud detection and credit scoring, for example. Another major use case is really two different types of simulation. Monte Carlo risk simulations show the promise for a quadratic speedup on quantum. Then there are simulations of the real subatomic world, such as materials science simulating molecules.
Cloud environments will have to mature how to share precious quantum resources efficiently for such real-time tasks to be feasible. For now, there are still queues, which means the systems are only appropriate for internal proof of concept projects rather than 24/7 production tasks.
Q. What boundaries should we be establishing around our use of artificial intelligence?
Andrew: There are a few key aspects of an organization’s exploration, adoption and ongoing use of the AI family of methods (including machine learning, natural language processing, computer vision, speech recognition and others) that companies looking to invest in AI should consider.
Strategy and governance is critical. The organization needs to define the strategy for the implementation and use of AI and determine whether that strategy has been vetted, approved and is aligned with broader organizational goals. This doesn’t require a lot of technical knowledge but is an important discussion among business leaders.
Then, consider whether the right resources and the right skillsets are onboard, to be able to effectively explore and drive to adoption and ongoing use and what that set of skills and resources looks like will differ depending on what aspect of AI is being explored and what way the organization is seeking to adopt it. Organizations that select a commercially available, third-party, OCR (optical character recognition) solution, for example, will have very different needs than a company seeking to build a custom AI solution.
Ethical AI is a discipline in its own right. What are we, as an organization, comfortable with letting a machine determine for us? Is it who gets a loan and who doesn’t? Is it the interest rate to be applied to that loan, is it who gets hired, who doesn’t, who gets fired, who doesn’t? And at what point do we want human checks and balances as part of that process?
Do we have the right processes and procedures? Can we say how we made this decision and why — this is the logic that was followed; was it rules-based or not rules-based. Can we reproduce aspects of that decisioning if required or requested? What underlying data are the models using and how do we know they are reliable and themselves don’t include or introduce elements of bias? What kind of platforms are we looking at making use of - are they proprietary or commercially available? Clearly, there are a lot of questions to be answered to ensure responsible use of AI.
There’s an opportunity to walk, crawl, run, fly here. Even if you don’t yet know how to spell AI, there are ways to get introduced to the subject and work your way through it rather than be terrified of it and do nothing.
Q) Will there be specialized coding or programming skills needed for quantum computing?
Scott: There are multiple programming languages developed to work with different hardware. There are also development tools that let you learn one language and then send the “job” to specific backend target quantum computers from different companies like IBM, IonQ, Rigetti.
Coders can’t just learn the language, though. At a minimum, they need a background in physics and linear algebra, plus knowledge of machine learning. They also need to be able to understand business problems and how to convert them to quantum algorithms. It’s a tough skill combination to find.
Andrew: Not just limited to quantum, I think that many functions need to look carefully at skills development and the broader acquisition and maintenance of skills on their teams. Specifically, look at skills around things like contemporary methodologies, agile, scrum, contemporary and emerging tooling. I spend a lot of my time working with internal audit leaders and internal audit functions. I think audit functions need to pursue a much broader and deeper knowledge base and set of skills, including keeping up with the latest trends around technology. Take cloud as an example. As internal audit functions think about their capabilities, how many would self-assess that they had advanced or expert level cloud skills on their team? Not many. It’s that kind of mindset, self-evaluation, and intent around skills acquisition, skills development and maintenance that I think is really a critical component of being prepared not just for the near term but also for the future.
Q. How can we prepare our executive leadership and boards to invest in emerging technologies?
Andrew: During the webinar, we discussed making sure that the business context specific to the organization, its strategy, its operations and its capability, is wrapped into the messaging. Introduce a basic understanding and appreciation of the technology to senior executives and the board in such a way that they can then make the right decisions. Do some information sharing and some briefing to stimulate discussion and generate interest; then revisit the topic and provide updates periodically. Share information about trends and themes, industry-specific or the broader business community and avoid getting into an unnecessary level of technical depth. Start using terms like bits and bytes, qbits, quantum entanglement and superposition and it’s not going to be a productive discussion. Avoid getting into technical jargon and detail; keeping it at an appropriate level of business and contextual discussion is important. Some of the subject matter can be highly technical but focus on key messaging that is relevant to executive level stakeholders, work to “educate up” rather than “dumb down” and engage executives and boards in the discussion.
To learn more about our emerging technologies consulting practice, contact us.