AI’s Bumpy Road to Fruition

Areas where AI is beginning to show returns are industrial automation (smart factories), decision support in medical diagnostics, new drug formulations (for example, COVID drugs), and business process automation such as fraud detection and interference with financial transactions.

In each of these cases, business value was easily demonstrated in operational savings, loss prevention, and speed of decision making.

It is success stories like these that make AI a compelling target for business leaders. Unfortunately, on a ground level, there is still a lot that needs to be done before most companies can take full advantage of AI.

Where most companies stand with AI

In March 2021, a Boston Consulting Group survey revealed that less than half of companies queried had mature software (AI) installed in production.

In some cases, business use cases are not fully defined, and organizations still view AI as an experimental technology. In others, use cases existed, but the organization was not ready to develop and install them. Barriers to readiness for implementation included a lack of high-quality data for AI use, as well as a general lack of preparedness for IT, data scientists, and users across the enterprise to fully deploy AI.

These were (and still are) hurdles to overcome:

1. Data silos

Companies still deal with data silos across their organizations that do not integrate with other sources and types of data in the organization. This will require developing a comprehensive data fabric that can link and weave all data into globally accessible data that everyone can access to remove silos.

Most companies don’t have these blanket cloths of data, so isolated data silos still exist, and no one can access all the data that can make an AI application truly informative. These limitations limit the ability of AI to produce high-quality insights that are completely reliable and immediately actionable.

2. Lack of AI tools

Data preparation, Transformation Extraction and Loading (ETL), business automation, intelligence software, and security governance tools are all essential to developing, deploying, and supporting a fully-fledged AI system in production. Many IT and data science departments still define these toolkits.

Most have not yet thought about what their software maintenance methods are for AI. Until the toolkits and procedures for maintaining the integrity of deployed AI systems are defined and implemented, AI will remain in the development stage.

3. Lack of people skilled in AI tools

The IT department needs to upgrade employee skills so that employees can effectively develop, deploy, and support AI. The life cycle of AI is an iterative one. AI testing comes within a certain percentage (say, 95%) of the accuracy of what subject matter experts in each discipline will conclude, so designing AI tests is very different from designing a QA script for DevOps or a traditional software application.

AI also runs on different operating systems and hardware than traditional software. The storage architecture of AI, which may need to store large amounts of data, must be structured between on-premises data warehouses and cloud data warehouses.

IT leaders will need to focus employee skills in these and other areas of AI.

The road to artificial intelligence success

In 2022, regulatory preparedness will be the main focus for companies working with AI, with one key caveat: In 2022, companies expect AI, analytics, and big data to deliver realistic results.

To deliver realistic business outcomes using AI, the IT department must be able to minimally check the following boxes:

  • Develop and deploy at least one business use case that “pays off” for the company by delivering faster and more reliable business processes that either lower costs or boost revenue.
  • Provide data and results that management trusts.
  • Develop AI methodologies and skills in information technology so that information technology can successfully develop, deploy and support AI.
  • Ensure strong AI security and governance.

Can IT do this?

In a November 2021 report, Gartner saw that organizations were still experimenting with AI and struggling to integrate AI into their standard processes. Gartner’s prediction was that it could take up to 2025 for half of organizations worldwide to reach what Gartner’s AI maturity model described as the “stabilization phase” of AI maturity.

If this prediction is correct, the key for IT leaders in 2022 will be to nurture AI into smaller use cases that they know will successfully demonstrate the value of AI to the CEO and other C-level executives. At the same time, CIOs must take steps to acquire Tools, building data structures, and developing employee skills that can support an imminent future for AI deployment on a larger scale.

What to read next:


Leave a Comment