You would think a company successfully advising global blue-chip enterprises on AI would have no qualms about this technology. Especially as no CEO or enterprise leaves out AI as a strategic imperative.
Far from it.
Tessella must have more highly qualified data scientists per head than any yet points out two major roadblocks that stall many an AI initiative. It asks two key questions.
- Can your AI cut it in real-world operations?
- Have you achieved user acceptance?
Cutting AI in real-world ops
It is easy enough to create a specific AI with data ready to feed into it. Fraud detection and prevention or Personal Injury Analytics- let's get rolling. But where do you put the output? How well have you trained the fraud or PI team to action the output?
Insurtechs with little domaine knowledge or your own data scientists in a lab environment often have little thought to production deployment. And once they tackle motor, how do they deploy to home and contents with different back-end core systems and data silos?
Make sure your AI strategy has this fully thought through- read the full Tessella article below- they know how to avoid roadblocks.
User Acceptance
A common source of unfulfilled AI expectations is user reluctance to support it: -
- Takes too long
- Too complicated
- UX is terrible
- I just don't trust it
Unless users were involved from the beginning they will probably mistrust the outcomes. Is it a real identification of cause and effect or just coincidental? Make sure they understand how AI reaches the predictions and decisions and how to leverage the outputs.
Running before you can walk is a common fault line when technology hype meets user scepticism. Again look at world-class practitioners like Tessella who have large enterprise-wide success and learn the lessons they did.
Take AI a step at a time. Some large insurers started with Fraud Analytics and or Personal Injury. With solutions that can access, join, analyze and make "AI Ready" the hidden data in silos and incompatible core systems inherited over the years through M&A.
Many have already sifted out fraudulent claims through an effective digital FNOL solution. Do you know that this is 35% for home insurers that have deployed 360Siteview?
AI is strategically important and for most insurers will take two to three years to deploy effectively. Read the Tessella article below and remember "Fools rush in where angels fear to tread".
AI capabilities are advancing apace in many enterprises, and AI is increasingly addressing many complex problems successfully. The challenge now is getting proven AIs out of the lab and into the enterprise. This requires forward planning and closer collaboration between data scientists, expert and non-expert users, and the business function. AI capability has come a long way. We must now ensure POCs are designed to work for the end user and are built to scale within that enterprise’s specific IT and human infrastructure.
unknownx500
