An unnoticed gem from the US Census Bureau has produced that rather startling figure. Usage is a bit higher in the information sector market, but it’s only 13.8%.
According to a survey, a further 6.5% plan to use AI in the next six months, but the wider cross-section of the markets is understandably more diverse and generally not moving much from the 2018 statistical base.
These are hardly mass-market penetration stats. The fact is and always has been that AI systems have critical cost considerations and constraints.
For instance:
AI systems have to integrate with a vast range of technologies and most importantly active business systems in real time.
Functionality needs to be mapped by any kind of business.
Cost benefits need to be clearly visible. These things can cost big money.
Commercial systems need to integrate with banking and other core business functions. This can’t be done on guesswork.
The chatbot function is largely irrelevant in many scenarios. Do you need a gossipy AI doing your accounts?
Big ticket acquisitions are tricky at the moment. The wider business sector is getting hit with rising costs, and interest rates, and so is the global client base.
A next generation of AI is becoming visible in the form of Artificial General Intelligence. The anti-human hype is excruciating.
AI systems are very new. All new technologies take time to wear in and make themselves fully operational.
Fixes for AI problems are being learned every second, but this is the very beginning.
All the hype about AI has generated natural distrust. That’s not a market sentiment that can be argued with; it needs proof of the value of AI.
Not everyone’s impressed with AI. Many people, including me, consider it clunky and as a patchwork of other older technologies. In some ways, it’s a “phone you can talk to”, rather than a major leap anywhere.
AI does have baseline uses. It can do reports, etc., but it can’t oversight itself. Judgment and interpretation are mixed blessings, too.
Another problem for business is “the illusion of efficiency”, particularly shedding staff. I know of a business that was said to have sacked all its staff and instantly left the boss to check everything herself.
AI definitely isn’t built-in expertise. It’s not knowledge. It’s automated information, based on other information of any level of quality and depth.
AI doesn’t relate to people in the same way people relate to people. That’s a huge obstacle in client-centric environments. Should an AI be infuriating a major investor? How about instant class actions based on its response?
The fact is that AI has a long way to go to deliver so much at the business level.
Caution is human nature, and that’s what these Census Bureau figures mean.
________________________________________________________________
Disclaimer
The opinions expressed in this Op-Ed are those of the author. They do not purport to reflect the opinions or views of the Digital Journal or its members.
