A Guide to Underwhelming with AI

  2 m, 59 s

This week in AI News & Insights: a guide to underwhelming with AI, featuring Watson for Oncology; why Forrester says 75% of early AI projects will underwhelm; and Stephen Hawking’s AI warning.

 A Guide to Underwhelming With AI

Jeff Catlin is not impressed by AI companies. Or rather, he’s very frustrated with companies who promise the world and fail to deliver. And Jeff wants to help you avoid those companies.

On that note, here’s Jeff’s latest Forbes article, “How to Underwhelm with Artificial Intelligence“.

Fact is, the AI field has a huge problem of over-promising and under-delivering. Remember “Watson for Oncology”? It was meant to help doctors diagnose cancer cases and make treatment recommendations. But a combination of biased data and dramatic over-promising led it to founder. Now, MD Anderson has officially “benched” the project, even after spending more than $62 million.

Watson for Oncology is not the only case of a major company underwhelming with AI. But it highlights the potentially disastrous financial impact. The question is, how can you avoid the same fate?

The first step is to understand what you’re dealing with. In this Forbes article, Jeff explains the common misperceptions of AI that lead companies to over-promise and under-deliver.

Next, read our list of 4 questions to ask before you choose an AI partner. And finally, contact us to discuss a custom, realistic “words-first” AI system to solve your unique business problems.

Why Forrester Says 75% of Early AI Projects Will Underwhelm

Scrolling through social media after reading that “underwhelm with AI” article, I stumbled across this tweet from Forrester:

“75% of early AI projects will underwhelm due to operational oversights. In 2018, leaders will need to reset the scope of #AI investments — and place their firms on a path to realizing the expected benefits.” – @forrester, 13 December 2017

Talk about timely! Turns out, back in December, Forrester released their technology predictions for 2018 – and some are already coming true. The report, ominously titled “Predictions for 2018: A year of reckoning”, describe a positive but rapidly evolving technology landscape in the year to come.

For example, Forrester points out that AI investments in 2017 focused on discrete use cases to deliver immediate value. Most of these technologies aimed to augment human intelligence or create conversational interfaces. But, says Forrester, because these projects focused on discrete use cases, the benefits will be short-lived.

Ultimately, Forrester expects that business leaders will reset the scope of their AI investments in 2018.

Click here to read a round-up of Forrester’s technology predictions for 2018 on Solutions Review

Stephen Hawking’s Different AI Warning

Amidst myriad tributes and memorials to one of the world’s greatest scientists ever, Mike Kaput over at Marketing AI Institute resurfaces Hawking’s slightly different AI warning:

“The real risk with AI isn’t malice but competence.” – Stephen Hawking, 2016 Reddit AMA

Hawking made this comment during his 2016 “Ask Me Anything”, hosted on Reddit. In this context, Hawking is talking about the risk of a “superintelligent AI”, such as the Cylons of Battlestar Galactica, or Terminator’s Skynet.

Now, superintelligent AI is a long ways away, but as Mike Kaput points out, Hawking has a good point. Many industries already use AI to perform cognitive tasks that would otherwise fall to humans. And it’s critical that we consider the implications and impact of new AI technologies as we develop them.

More Weekly AI News & Insights

Stay tuned to our blog for more weekly AI news and insights articles, interest pieces and thought leadership.

In the meantime, why not learn how Lexalytics can transform your text data into insights and money? Drop by our website for more info.

Categories: Newsletter