NEW articles

The Art of Change Management in IT projects

In “Leading Change” (1996) John Kotter observed that 30% of organizational changes fail. 25 years later, McKinsey reports the same 30% failure rate for enterprise systems projects. Why change management is still an afterthought, and what can we do about it? > Read more


CIO In Five Index Cards

Last year I saw a nice little book titled “The Index Card: Why Personal Finance Doesn't Have to Be Complicated”. That gave the idea to write this short post: the CIO job in five index cards. > Read more

Every business can use satellite imagery

One of the most powerful applications of AI, the exploration of satellite imagery, is accessible to almost any business. Launched 80 at a time, the low-cost “cube” satellites stream data to AI models that mix image recognition and analytics to document the world. Together they give access to a detail of information at a speed and cost unthinkable a few years ago.  > Read more

The master algorithm

Artificial intelligence is a terrible name for this business. The picture of a proto-human robot comes to mind. The reality is computer programs having nothing to do with each other. Still, the perception is that machines are growing smarter by the day. Each of them in its own way, but can we imagine to a convergence?  > Read more

How AI can help against on-line fraud

When fraudsters use camouflage to mimic human behavior and escape detection, old-school methods to find them are no longer enough. Online merchants are stuck between the need to be nimble - guest checkout and 2 days delivery - and mounting losses. Increasingly they look at machine learning services as an answer. Scanning billions of transactions for anomalies seems to be right up in its alley. But how far are we from protection as a service?  > Read more

How AI uses emotions

Many AI teams include psychologists to help with emotional design. AI combines voice and facial recognition with data mining to assess how we feel, predict what we are likely to do, and adjust the response. The question becomes who is using the results to do what?  > Read more


Starting AI projects
the easy way

"It's time to retire the priesthood" wrote Frank Chen: AI is not complicated. It might be the new electricity but to get started we can use it as the new database.  But the daily news on the start-up scene or research that assume a solid level in calculus is not too much of a help.  What is the easier way? What are the real obstacles?  > Read more

AI Potemkin Villages

AI "black box" issue had made headlines in the most serious publications. The doubt is:"why should I trust a system I don't understand". The answer maybe be less technical than social ...  > Read more

Five pillars of an AI Strategy

Only 11% of the enterprises recently surveyed by Gartner report that understanding AI is an obstacle. The real ones are finding the right people (54%), good use cases (37%) and to build a strategy (35%). The challenge is not why but how. An AI strategy can start with five questions. > Read more

AI as enterprise innovation

Many components of an AI strategy are in the CIO domain, but creating the AI team within IT may reduce its impact. A better model is to set AI as an Enterprise innovation, using the scientific toolbox ...  > Read more