Every 20-30 years, a technology breakthrough so dramatic transforms the enterprise and our lives. The next pivotal moment is right now.

Generative AI – specifically Large Language Models (LLM) like GPT popularized by OpenAI and Microsoft, have catapulted us into a new era by democratizing AI for enterprises. A historical analogy is how the advent of web browsers such as Netscape Navigator in the early 1990s became the catalyst moment which triggered a chain of events changing the way we work, connect, consume, collaborate and live. With browsers, everybody could engage via point-and-click. Internet evolved from being a platform for, and by tech enthusiasts into a platform of mass adoption and mass change. Browsers democratized the internet. Generative AI is democratizing AI today in an analogous way. What was once available to only a handful of companies with access to massive supercomputing power and billions of dollars of yearly spend, is now available for all to customize and consume as a pervasive ‘copilot’ that guides us as we work, consume and live.

What happens next will be astonishing and transformative. This is the catalyst moment which is making AI become the main computing platform for 2023 and beyond. General Purpose AI is obviously not here yet. But we are seeing AI as the next mainstream computing platform, ushering us into a world where enterprise leaders need to adapt an AI-first mindset.

How did we get here? A part of the answer are the steady advances in machine learning and deep learning since the 2000’s. The other part of the answer is the recent accelerated transition to the cloud. As a result, a critical mass of companies is now partially in the cloud with soon-to-be good-enough data platform and governance. These organizations are now in the process of evolving their focus of journey-to-cloud to extracting value-in-cloud. This quest for clear business value is intersecting with AI’s ability to deliver such value, creating an accelerating and pervasive AI-first world. AI-first is not a solution looking for a problem to solve, but rather the ability to rapidly address real opportunities and create durable competitive advantage for those who will adopt wisely.

How impactful is AI really going to be?

We believe it will trigger massive waves of change which may redefine entire industries and unleash a new era of GDP growth. It will impact the sheer ‘fabric’ of many companies and change their operating models as the cost to take idea creation to idea instantiation will be approaching zero. It will redefine the relationships between customers and businesses, and between citizens and public organizations. It will certainly cause businesses to rethink how products and services are created and how they go to market. It will shift the way business leaders think about the customer experience and how office and frontline workers can be empowered and supercharged. Also the scope and nature of IT departments will be redefined—right down to how software is written, how applications are being created or modernized—even how their enterprise cloud is being managed and operated.

Here are 3 reasons why we are so bullish about the impact of AI:

AI will empower organizations to tackle difficult problems that take advantage of the world’s knowledge across company boundaries, and within the proper legal frameworks. Imagine AI reasoning across the knowledge corpus of companies that traditionally have worked together, may have shared data, but not practiced ‘shared cognition’. For example, a manufacturer who creates a new material for tires because of multiple shared knowledge corpuses across the manufacturer, government, a chemicals company, and the investment arm of a financial services company. AI-First is the growth engine for multiparty systems greatly extending the data footprint and capability of a single organization. And instead of just storing and analyzing data like many companies do today, these new cognitive AI models will keep the data “alive”, creating a constant feedback loop (but also a need to manage the risk of more “digital noise").
AI will force a rethink of the relationship and engagement between people and technology, perhaps as fundamental as during the Industrial Revolution. AI-first is inherently people-first, impacting everyone and blurring the distinction between ‘knowledge workers’ and ‘blue collar workers.’ The level of machine cognition is now good enough that we can tap into the vast “librarian” of the internet and combine with hard-to-navigate “locked away” enterprise knowledge stores for creation of new content or seeking/finding answers. The notion of having a copilot or assistant is realized, revolutionizing how we think about task completion and business process reinvention. AI decisions will however need to be uniquely woven from a company’s values. Successful leaders will be the ones who deeply understand how to leverage AI to support employees’ creativity and quality, and share lessons about how employees can use these models to strengthen their sense of workplace contribution and value. We will need to ensure that we don’t create new distinctions between ‘augmented’ and ‘non-augmented’ workers. We will perhaps need new team structures and new organizational models.
Speed and Scale. Speed is not just about computer chips being 100.000X faster than electric signals in our brain (although they are!). It is about the vastly accelerated time-to-value we’re already seeing in the initial real-world use cases. Looking ahead, we believe this acceleration will expand from discrete AI projects to encompass the rhythm of business itself as creation and intelligence will occur at the speed of computers and at the hyper-scale of cloud. Another dimension is the intense competition (and cooperation) in the ecosystem, and the learning-by-doing-at-internet-scale which is accelerating innovation. The limitations of today may seem trivial tomorrow.

The waves of AI innovation will be swift and simultaneous

AI will trigger continuous ripples of change in several overlapping waves of innovation over the next 48 months:

0-12 months: Many sites and apps that already have a freeform text field will make AI even more mainstream and user friendly when they become powered by a large language model. AI as an API becomes just as ubiquitous as accessing a database, going beyond text, to code, designs, images and video. This will be as common as spellcheck is today. Everyday users will have the power to build increasingly sophisticated applications that leverage the broad capabilities of LLMs.
0-24 months: More and more AI capabilities will be continuously and rapidly embedded into line-of-business applications and modern workplace tools. An example may be risk detection and mitigation for a supply chain, where a delay due to a complex mix of internal and external factors – like a strike or weather – will be anticipated, a mitigation plan automatically created, then communicated to stakeholders and enacted. The nature of work will also evolve. Copilots will uplevel and democratize skills, with experienced workers of all types becoming more creative, and beginners becoming more productive.
Insights buried in mountains of data or hidden in application silos will also be unearthed faster. For example, an oil and gas company with AI-driven knowledge management will change the way it addresses the issue of “How do I fix a problem with pipeline TYS7729-1 at location 78.1493 where valve A651 is stuck due to corrosion?” Generating an answer and a repair plan will take minutes and hours, rather than days and weeks of complex teamwork.
Increasingly, we will see industries pilot transformational scenarios. For example, in banking, AI may empower a new breed of consumers to be their own financial advisors. Manufacturers may use AI to create new materials (i.e., corrosion resistance) with generative design assistance vs. the ‘ah-ha!’ serendipity of R&D.
18-48 months: AI upends existing business models and new ones are created as the art of the possible is redefined as the cost of creation is approaching zero. Some will use AI just to “breed faster horses”, but the leaders will “invent the automobile.” The question they will ask is not how can we automate processes just to reduce costs, but rather what can be created and how can technology and human ingenuity be empowered for every person and every organization?

How can organizations begin to adopt an AI-first mindset now?

Organizations are facing the difficult reality of transforming to respond to the unpredictable market forces of today, while also preparing to perform and grow in the future. As they navigate this new era, we believe five key factors will help organizations on this journey:

People are, and will always be, critical. For AI to be a successful assistant, we will all need help to learn how to be comfortable working iteratively from generated concepts that need to be tweaked, refined, enriched and approved. On a more fundamental level however, the bigger opportunity will be to achieve the promise of the future of work where humans flourish. The best will take a step further and ensure that AI contributes to social goods like education, financial opportunity, and healthcare, rather than simply an increase in digital noise. This is planetary-level change enablement leading from the front.
Be clear on the business “why”? A consequence of the shifting focus of journey-to-cloud to extracting value-in-cloud is that Boards today have limited strategic patience. Hence the healthy need for being laser focused on immediate value. In the short term, we see the most common scenarios being: 1) improved customer sales and service; 2) operational cost reduction with document/image generation and process automation; 3) real-world employee productivity with enhanced knowledge management and routine task automation; and 4) better/faster software development. We recommend selecting the initial use case where these capabilities can be integrated in a modular, yet scalable, way and ideally where you already have an existing capability, like an existing chatbot. You should also start where the data is qualitative and diverse, because…
…it all starts with data. As our colleagues in Accenture point out, it is critical to think about data as having the potential for intrinsic value if treated as a product that can unlock its dormant value. This requires the breaking down of data silos, removing duplication, creating trusted data products, reducing the cost of data rework. Perhaps most importantly, it requires the democratization of access by more widely sharing high-quality data products across the ecosystem.
Trust is paramount. That means trust in technology, but also trust through technology. These changes are having a tremendous human impact, and the potential for unintended consequences will create ethical dilemmas that the world does not yet have robust standards or legal precedent for. It’s not enough to repeat the mantra of “do no harm”. The right answer is to programmatically start embedding digital ethics into your organization and processes today. Key topics to focus on are: 1) robust, ongoing risk-based oversight; 2) respect for intellectual property; 3) solid information security and privacy; 4) transparency and explainability; 5) sustainability; and 6) intentional sourcing.
Experiment courageously and wisely. Adopting an AI-first mindset won’t happen overnight, and we may not know what we don’t know. Sometimes use cases can only be discovered by experimenting. We’re certainly not promising to know it all, but we can help you take some practical actions to move your organization and people forward.

As a starting point, we invite you to register for a complimentary two-hour workshop to explore opportunities to drive business value from AI in your organization.

Making the shift to pervasive AI: Adopting an AI-first mindset