AI: Failed Promise Or A Case Of Unrealistic Expectations?

Following bold claims about the business benefits and even world-saving power of artificial intelligence (AI), it’s not surprising that talk of broken promises is growing. However, has AI failed to deliver, or are we setting our expectations too high?

Recently, consternation has centered around Covid-19 as some initial AI-assisted breakthroughs quickly fizzled out. For many, frustration stretches back much further, with a 2010 study showing 38% of organizations lacked the understanding of how to use analytics to make better and faster decisions. Today, fewer than 25% of global organizations have developed an enterprise-wide AI strategy. Yet as companies continue pouring more dollars into AI tools, it’s increasingly important to establish whether investments are worthwhile.

Putting the hype aside, AI does have significant potential to help organizations enhance their efficiency and make better use of data. However, to use it effectively, one must be realistic about its capabilities.

AI Expectation Versus Reality

Amid impressive predictions, inflated expectations of AI are easy to understand. PwC projects that AI could bring $15.7 trillion to the global economy by 2030, and the Harvard Business Review notes that subsets such as machine learning are set to bring gains of up to $2.6 trillion for marketing and sales. However, implementation alone isn’t a fast-track to efficiency.

In a recent Capgemini survey about customer experience automation in financial services, companies saw quick wins from using chatbots, including lower costs. However, tools didn’t deliver where it mattered, with 49% of customers seeing low or no value from interactions. This example illustrates the sizeable risks of viewing AI as a magic solution and overlooking practical realities. As noted by frequent Harvard Business School lecturer Mark Esposito, going beyond proof of concept is proving difficult for organizations, as they get caught up in lofty visions and find they “can’t convert their idea into code and algorithms.”

Making the most of AI will involve identifying where it can add value, what’s needed to make it work and how companies can build the necessary infrastructure.

Getting The Basics Right

To make wise investments in AI tools, businesses require an honest view of current data processes, people and technology. This marks a company’s initial steps toward establishing the fundamentals of data maturity, which are critical for maximizing AI’s value.

First, companies must look to build an insight-driven culture. This means placing the insights from your AI at the heart of everything you do. Schedule regular meetings to review what your datasets and AI are telling you — and then take actions based on them. Too often, AI insights are used to justify existing decisions rather than drive new ones. Staff should likewise be encouraged to use the data and insights available to them without fear of failure.

To do this requires educating staff on how to use data and AI tools effectively. Businesses can start that process by embedding data expertise across all departments. However, leaders should start recognizing skills gaps across all staff and addressing them with professional development. Scaling educational initiatives for different audiences — such as developing training modules applicable to specific roles or accounting for different learning styles with a combination of in-person training, on-demand videos and online documentation — will help businesses make the most of their chosen technologies.

By establishing this foundation, businesses can then review their ways of working, assess their current challenges and identify which solutions might help overcome them. If dealing with data fragmentation, for instance, they can enhance visibility and usability by integrating information from all sources. When all relevant data is collected in one place, teams can better leverage AI tools to strip it of its various formats and analyze it in real time. For each challenge, businesses need an in-depth understanding of how technologies can support their needs in a coherent and cost-efficient manner.

By evaluating all these business elements, companies will be better positioned to tap AI for many useful — though not necessarily revolutionary — business purposes.

The True AI Value Proposition 

Most realistic AI use cases revolve around data, particularly cutting time to insight. With data volumes on track to hit 74 zettabytes globally, traditional manual preparation is taking longer and insights are arriving later. The shift from reactive coding-based to visual analytics has increased accessibility and reduced production from weeks to days, but ongoing feature adjustment can only minimize waiting time so much.

Moving toward AI-supported augmented analytics can generate actionable insights within minutes and a host of business benefits, including:

1. Regaining data control. Transferring the heavy lifting of data preparation to smart machines can dramatically accelerate integration, harmonization and enrichment, delivering fast access to unified insight. In addition to saving manpower, this allows for greater collaboration and shared oversight between teams, allowing businesses to deliver value faster.

2. Uncovering hidden gems. On top of extra processing muscle, AI brings deep-diving capacity for augmented analytics. Algorithms can constantly monitor incoming data for hidden anomalies and trends that would be missed by humans. This opens opportunities to be more responsive to constantly changing market conditions and trends instead of trawling through data manually.

3. Optimizing insight agility. Where automatic notifications highlight important patterns, smart predictions make real-time recommendations that enable companies to get ahead. For example, assessment of social media interactions might reveal signs of where consumer attention is shifting, which can then inform product development.

Few tools can promise effective transformation straightaway, and AI is no exception. However, that doesn’t mean it should be chalked up as a failed dream. AI-powered augmented analytics can already give businesses the means to wield data, extract previously obscured insights and look ahead to pinpoint what steps they should be taking next. Used in conjunction with integrated and reliable data, augmented analytics is a discovery journey that can deliver tangible long-term value.

 

Original post: https://www.forbes.com/sites/forbestechcouncil/2021/08/09/ai-failed-promise-or-a-case-of-unrealistic-expectations/?sh=1c755d6e312d&s=09

Leave a Reply

Your email address will not be published. Required fields are marked *