A strategy for leaders to implement AI-driven change
The pace of advancement in artificial intelligence (AI) continues to accelerate rapidly.
In just the last few years, we have seen major leaps forward in AI capabilities driven by innovations in deep learning, neural networks, and vast improvements in computing power. Milestones such as DeepMind's AlphaGo mastering the complex game of Go, AI systems beating top humans in multiplayer video games, and significant gains in computer vision through deep learning have showcased how sophisticated AI systems have become. Natural language processing also continues to advance, with AI assistants like Siri and Alexa handling increasingly complex voice commands and queries. In robotics, AI is enabling greater autonomy, adaptability, and mobility. In generative AI space, growth of chatGPT was like an "iPhone moment". Together, these and countless other advances across all facets of artificial intelligence are taking us swiftly towards the emergence of more general artificial intelligence, with revolutionary implications for how humans live and work alongside AI.
AI capabilities are achieving new heights across domains, presaging industry disruption. Savvy leaders closely track AI progress, strategically integrating innovations ahead of the curve. They understand that successfully capitalizing on emerging opportunities requires proactive adoption before disruption strikes. AI waits for no one – technologically adept leaders actively shape its trajectory.
To actively lead their organization's AI future, forward-thinking leaders should focus on these seven essential strategic building blocks:
1. Be Agile:
I am not specifically recommending the adoption of an agile development framework. Traditional waterfall software development workflows may still be appropriate in many circumstances. However, when evaluating and implementing generative AI technologies, it is advisable to proceed with an "agile" mindset, moving rapidly to assess and incorporate these innovations while remaining mindful of potential challenges. Leaders should aim to make swift, informed decisions around if and how to leverage generative AI capabilities in a manner that balances benefits and strategic alignment with prudent evaluation of risks and responsible mitigation planning.
2. Use cases, Use cases, Use cases:
You, as leader, should take a clean-slate approach to identifying where generative AI can transform operations and unlock new sources of value. Seek out use cases across the business where intelligent automation can boost efficiency, accelerate growth, enable new business models, and open up revenue opportunities. But pursue this AI-powered innovation through close collaboration between technical and business partners - tap cross-functional insights to deeply understand user needs, business priorities, and desired outcomes. When evaluating potential AI implementations, carefully consider total costs across build vs buy decisions, deployment, maintenance, and model degradation risks. Also factor in less tangible costs like change management, employee training, and potential job impacts. Approach AI judiciously, not just chasing the new and shiny.
See what I envision telecom industries to look like after they apply some of the generative AI use cases.
3. Build vs. Buy:
This dilemma has long vexed technology leaders when evaluating new solutions. This same deliberation applies to generative AI. Most of the generative AI capabilities are run by foundation models and you can use one provided by 3rd parties (I am biased to Amazon Bedrock but you can also look into other providers). Alternatively, you can build one from scratch (like Bloomberg did it) or build on top of an existing open-source models (as of this writing, LLaMA 2 and Falcon models are leaders). Decision to build vs. buy will boil down to 3 things: Time, Cost and Skills, but you, as a leader, should create a guideline for reliability, security and bias when buying or building on top of open-source models.
4. Generative AI part of your technology strategy:
Though generative AI is revolutionary and it can solve almost any use cases, it is imperative to understand that generative AI doesn't operate in vacuum and it does perform better (and improve margin!) when used with other technology investment you have already made. Consider evolving your technology architecture to enable enterprise-wide adoption, interoperability, and responsible scaling of generative AI capabilities. If requires, modernize your tech stack to seamlessly orchestrate generative AI alongside current AI/ML models, apps, and data.
5. Data, Data, Data:
There is merit to the analogy that data represents the "new oil" of the digital economy. To fully capitalize on this digital oil, one requires a powerful engine - which generative artificial intelligence provides. Your company possess proprietary data which is meant for that particular industry. If that's not properly transformed to integrate with generative AI then you need to take the initiative to make it high-quality that will realize significantly greater business value than those relying on generic data. Don't trust me? Take a look at the results from Bloomberg GPT.
To maximize competitive advantage, consider building modernized data platforms ready to fuel generative AI and machine learning initiatives. With elevated focus on curating high-quality training data and data management fundamentals, you should realize the full potential of your "digital oil" and generative AI to drive transformative capabilities, performance, and insights in an ethical manner.
6. People:
While some early adopters may be well ahead in learning about generative AI, remember the vast majority are still trying to understand its impact on the company and their own roles. As a good leader, invest in training your people for new skills – your data or business analysts may no longer need to do manual information capturing if models can automate it. Instead, their role could morph into "prompt analyst." Be prepared for shifts like this and guide your teams through the changes. Existing employees are your best asset for moving forward in the AI era with the right skills and mindset.
7. Risk and Governance:
Like any new technology, generative AI carries risks - whether around data privacy and security, biased models or data leading to skewed results, toxic inputs generating harmful outputs, insufficient fairness or transparency, or lack of governance over the AI pipeline. However, consider prudent mitigation of downsides through governance and oversight. That will provide immense opportunities to enhance productivity, innovation velocity, and human achievement using AI-powered automation and augmentation . Generative AI is quickly evolving from a tactical tool to a strategic imperative. The window to gain advantage by embedding it into your technology strategy, workforce, and operations is open now. Act swiftly while balancing the risk.
To summarize, you all are farsighted leaders who are continually staying abreast of artificial intelligence advancements, adopting innovations ahead of the competition. With technological acumen and strategic foresight, you will proactively integrate AI capabilities before market disruption arises. You know that realizing the full value of AI requires shaping its trajectory within their organization in collaboration with cross-functional team, identifying the right data and implementing right risk and governance measure. You know how to invest in your people and when to re-skill them. AI's potential is not going to wait for anyone and those leaders wise and agile enough to capitalize on the possibilities will be rewarded.
