AI Needs an Enterprise Operating Model for Success

Power of AI

For today’s enterprises across industries, Artificial Intelligence is not only a game-changer but of critical importance to stay relevant and forge ahead. AI not only improves business processes and outcomes but also helps with better decision-making abilities, service offerings, monetize data and create brand value.

But just the idea by itself is not enough. The need of the hour is a robust Enterprise Operating Model to ensure success.

Confusion around AI in the Industry

AI is emerging as a disruptive field but also suffers from lack of consistency in understanding, terminology, usage and potential of AI within the Industry, including business and IT executives, AI practitioners, suppliers, integrators and vendors.

• AI is new,incorporates newer technology concepts and still somewhat difficult to understand for executives.Business executives and AI practitioners differ in their definition, understanding and potential of AI
• Is Predictive Analytics a form of AI? How about rules-based automation aka RPA/RDA/DPA?
• How does Machine Learning and Deep Learning relate to AI?
• Many ISVs and Start-ups mistakenly market products as AI-enabled/ AI -powered without enough qualification
• Lack of understanding of foundational aspects of AI (e.g. data strategy) that underpin longer term effectiveness
• Plethora of AI technologies adds to the confusion, some will mature fast while others may fall-off so risky bets currently (ref: Gartner Hype cycle)
• Newer technologies/ applications are constantly being added and standards are changing so the area will remain fluid for a while
• Differing analyst views on dealing with AI and handling challenges especially the timing and extent of disruption
• Emerging issues around ethics and governance of AI
• Which role/executive should own AI within the enterprise? Would there be single or multiple owners? What organizational model may make sense?
• Organizational reluctance to progress on AI given potential for displacement of jobs, or conversely where there’s executive mandate there’s a rush to progress AI on many fronts in totally unmanaged ways

“Just as electricity transformed almost everything 100 years ago, today I actually have a hard time thinking of an industry that I don’t think AI (Artificial Intelligence) will transform in the next several years.” ~Andrew Ng, Computer Scientist & Statistician

Initial Experiments and Path to Scalability

Broadly, the industry is either exploring or experimenting with AI technologies with isolated, narrow use-cases in individual departments/ functions or LOBs. These are conducted either by skunkworks teams or supported by product-vendors working directly with the business with limited involvement of enterprise IS. While these initial prototypes may show promise, scalability for broader adoption will need several concerns to be addressed.

Risks around AI and Value of Controls – Ethics, Trust, Privacy

AI is ‘amoral’ with no inherent bias for morality or ethics. AI can also be manipulated to maliciously cause harm to individuals, organizations or society. There can be unintended negative consequences from AI as well.

“I’m increasingly inclined to think that there should be some regulatory oversight, maybe at the national and international level, just to make sure that we don’t do something very foolish. I mean with artificial intelligence we’re summoning the demon.” —Elon Musk warned at MIT’s AeroAstro Centennial Symposium

Organizational Culture Change for AI

AI will need new forms of ‘Cultural leadership’ that values developing agile mindsets, engaging with workforce, and iterating rapidly vs cascading top-down changes. AI should ‘amplify’ human ingenuity with intelligent technology to make jobs more fulfilling but people need to continue to re-skill and broaden their knowledge and skill-base. The business environment will continue to be fluid with frequent disruptions that needs nimbleness to work across organizational boundaries.

“It’s natural to wonder if there will be a jobless future or not. What we’ve concluded, based on much research, is that there will be jobs lost, but also gained, and changed. The number of jobs gained and changed is going to be a much larger number, so if you ask me if I worry about a jobless future, I actually don’t. That’s the least of my worries.” — James Manyika, Chairman and Director, McKinsey Global Institute (MGI)

Workforce Displacement and Rebalancing

Workforce displacement caused by AI deployments needs to be planned out ahead of time as part of the AI program. HR, Internal Training and external partners need to work together to continue to re-skills, up-skill and cross-skill impacted employees. A culture of agility, flexibility, and adaptation is key to success as organizations re-balance their workforce continually. Cultural change, workforce engagement, learning, re-skilling, and job re-balancing is a core part of AI programs and cannot be an afterthought since that will compromise operational effectiveness and value from AI.

• Deloitte’s “2018 Global Human Capital Trends” report showed a lack of confidence in HR to address displacement. It found that while 72 percent of respondents think adopting AI is important for their business, only 31 percent feel ready to address it. HR’s lack of fluency in using analytics to improve workforce effectiveness is a factor.
• 36 M Americans have high exposure to automation, and upwards of 70% of tasks done by humans can be performed by machines across industries. This is likely to be a reality within the next decade. Economic factors will play a role, so economic downturns will be particularly severe for jobs.
• Distinction between jobs and tasks constituting the job is important, as tasks become automated not necessarily jobs, at least in the near term. Roles will be reconfigured as tasks are either replaced or augmented by technology.
• Broadly, Gartner has predicted AI will create more jobs than it eliminates throughout the economy although it will be disruptive for individual companies or industries.
• Careers need to be built around learning and not just existing sills. This is a crucial cultural change that needs to be embraced by all and workforce rebalancing programs need to facilitate this change.
• AI itself could be used by companies to help retrain (and retain) workers. AI driven training programs can deliver personalized training.

Exavalu’s Enterprise Operating Model for AI

Exavalu’s Enterprise Operating Model for AI is a holistic framework that considers all aspects to mature AI adoption in ways that improve business value & outcomes, adapts to newer technologies as they evolve, integrates with value chain participants to accelerate AI, fosters open collaboration with alliance partners, addresses talent gaps, organizationally centralizes certain functions yet federates others, and provides ways to handle culture change and employee rebalancing.

Organizational Appreciation and Business Value –

• What is the level of appreciation between/amongst business executives and practitioners around AI?
• What campaigns are needed to align stakeholders to achieve organizational momentum?
• What use cases are in line with the organization’s vision and plans?
• What use cases are at the right level of complexity and risk that are right for an organization’s journey? What could be the success criteria?
• What AI initiatives have been undertaken so far and with what results?

AI Technologies –

• What technologies have been invested in and with what results?
• What platforms (if any) has an organization has committed to that have AI capabilities? What high-compute infrastructure?
• Does the architecture allow modularity and standard integrations?
• What’s the organization’s experience with Low/No Code platforms?

Data Strategy and Platforms –

• What’s the organization’s strategy around Data? What is the roadmap and assets already in place that can be leveraged?
• What data sets, data sources, data quality, data catalog, high value data?
• Data Lineage and systems of record?
• What data privacy and security models are already in place?

AI Operating Model –

• How is AI being delivered currently and with what learnings?
• What is the Talent Strategy for scarce AI talent? How are partners being leveraged?
• What areas can benefit from centralization? What product management discipline exists for ongoing refinement?
• How is AI being operationally implemented into business processes? What gaps exist?
• What standardized training, toolkits, methodologies exist and are needed?

Culture Change and Workforce Displacement –

• How can open, collaborative, learning culture be introduced if already not in place? What is the composition of the workforce that will have to be realigned? Who are the AI Champions that can drive adoption?

Trust, Ethics, Governance and Controls –

• From initial experiments, what challenges have surfaced? What are the organization’s ideas around establishing trust, handling ethics and instituting governance and controls?

Reference:

• Gartner Hype Cycle for Artificial Intelligence, 2019
• AI, robotics, and automation: Put humans in the loop 2018 Global Human Capital Trends

About the Author:

Jayati Hazra is a Director with Exavalu and has 20 years of experience in Enterprise Solutions, Business & Technology Consulting in Lifesciences, Public Sector and Manufacturing industries. You can reach her at Jayati.Hazra@exavalu.com