Cutting Through the Hype of AI
- By CDOTrends editors
- January 24, 2020
Most companies struggle with how to successfully integrate artificial intelligence (AI) into their businesses, according to a new report on the state of AI and how companies can improve decision-making around AI.
AI can be a gamble
Titled: “Artificial Intelligence: A Framework to Identify Challenges and Guide Successful Outcomes,” the report by advisory firm Lux Research looks at the current state of artificial intelligence and offering some tips on how organizations can make their AI projects a success.
Of note is the assertion by the report that integrating AI can be a gamble for many companies. Decision-makers often do not fully understand the technology and have not thought through the true costs of implementing AI in their businesses. This culminates in marketing messaging being used to fill in the gaps where technical understanding leaves off.
To combat that, the report cautioned against investing too early in applications that require more advanced AI capabilities, as they are often too immature and untested.
“[In the past] researchers used everything besides the term AI to describe what they were working on; today, everything from a grammar checker to a stock trading application is called AI,” wrote Cole McCollum, Lex’s lead analyst for the report.
“Only then can companies make judgments on the maturity level of any particular product and fully understand the challenges inherent in implementing and maintaining these cutting-edge technologies. No investment is risk-free but following this framework will help avoid costly mistakes,” he explained.
Getting AI to work
To increase the chances of success, he advised businesses to consider the following four factors to make the right AI investments and decisions.
- Clearly understanding the outcomes implementing AI will provide for their business
- Focusing on an AI product's capabilities instead of flashy marketing;
- Knowing when the technology is mature enough to mitigate risk;
- Identifying practical challenges to both implementation and maintenance of the technology once it is in place.
A framework should be established to cut through “buzzworthy marketing” and to help businesses accurately assess AI’s capabilities based on their unique business requirements, suggests McCollum.
Ultimately, it is clear is that the hype around AI will remain, and likely increase, in the coming decade. This means that a sober approach to AI investment that prioritizes company needs and ROI and avoids a “technology-first” approach will ultimately be better-positioned for success.
The executive summary for the report can be downloaded here (pdf).
Photo credit: iStockphoto/Gerasimov174