A few years ago, the most popular buzzword in the business technology world was “big data.” This refers to the collection of large amounts of information in an organization that can be used to suggest previously unknown ways of operating or to surface ideas about what strategies are best to pursue.
What is becoming increasingly clear is that the challenges that businesses face in leveraging big data to their advantage remain, and that a new technology, AI, is bringing them back to the surface. Unless we address the issues plaguing big data, AI implementations will continue to fail.
So what are the issues holding us back from delivering on the promise of AI?
Most of the problems stem from the data resources themselves. To understand this issue, consider the following sources of information used on a very average workday.
For small businesses:
- Sheets stored on users’ laptops, Google Sheets, and the Office 365 cloud.
- Customer relationship manager (CRM) platform.
- Email exchanges between colleagues, customers, and suppliers.
- Word documents, PDFs, web forms.
- Messaging app.
For enterprise businesses:
- All of the above, plus
- Enterprise resource planning (ERP) system.
- Real-time data feed.
- data lake.
- Disparate databases behind multiple point products.
Please note that the above simple list is not comprehensive, nor is it intended to be. What this tells us is that there are about 12 places where information can be found in just 5 lines. What big data needed (and probably still needs), and what AI projects are also based on, is to somehow bring all these pieces together in a way that computer algorithms can understand.
Marketing giant Gartner’s 2024 Hype Cycle for Artificial Intelligence estimates that AI-enabled data will be on the upward curve of the Hype Cycle, taking two to five years to reach a “productivity plateau.” Given that AI systems mine and extract data, most organizations, except the largest, don’t have the foundation to build on, and AI assistance in this endeavor may be another 1-4 years away.
The fundamental problem with AI implementation is the same as when big data innovations have gone through hype cycles in the past. From sparks of innovation, rising expectations, valleys of disillusionment, slopes of enlightenment, and productivity plateaus, data comes in many forms. May be contradictory. Perhaps they adhere to different standards. May be inaccurate or biased. It could be highly sensitive information, or it could be old and irrelevant.
Transforming data to make it AI-ready is still a process that is more important than ever (perhaps even more so). Businesses looking to get a jump start can try out the many data processing platforms currently available. You can also start with individual projects as testbeds to assess the effectiveness of emerging technologies, as has become common advice.
The advantage of modern data preparation and assembly systems is that they are designed to prepare an organization’s information resources in a manner designed for data used in AI value creation platforms. For example, you can provide carefully coded guardrails to ensure data compliance and protect users from accessing biased or commercially sensitive information.
However, the challenge of producing consistent, secure, and well-formulated data resources remains an ongoing challenge. As organizations acquire more data in their daily operations, compiling the latest data resources to utilize becomes an ongoing process. If big data is considered a static asset, data for AI ingestion must be prepared and processed as close to real-time as possible.
Therefore, the situation remains a three-way balance of opportunity, risk, and cost. Vendor and platform selection has never been more important for modern businesses.
(Source: “Inside the business school” by Darien and Neil is licensed under CC BY-NC 2.0.)
Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo in Amsterdam, California, and London. This comprehensive event is part of TechEx and co-located with other major technology events. Click here for more information.
AI News is brought to you by TechForge Media. Learn about other upcoming enterprise technology events and webinars.
