A timely reminder that in addition to having a vision and strategy that defines the role of AI there needs to combat bias.

"I'm not concerned about AI super-intelligence “going rogue” and challenging the survival of the human race – that's science fiction and unsupported by any scientific research today. But I do believe we have to think about any unintended consequences of using this technology."

Richard Socher, Chief Scientist, Salesforce.com 

Read how to anticipate and avoid bias in his article. There is also a valuable review by McKinsey based on banking world and relevant to insurance and other sectors.

Read: Derisking machine learning and artificial intelligence

And remember; under GDPR enterprises can only use data for the purpose agreed by the data owner. Google and the NHS have already run into trouble over this. Easy to do when you implement AI for uses other than that which the data owner agreed to.