Recent advances in artificial intelligence (AI) and machine learning (ML), such as computer vision, are now available as intelligent services and their accessibility and simplicity is compelling. Multiple vendors now offer this technology as cloud services and developers want to leverage these advances to provide value to end-users. However, there is no firm investigation into the maintenance and evolution risks arising from use of these intelligent services; in particular, their behavioural consistency and transparency of their functionality. We evaluated the responses of three different intelligent services (specifically computer vision) over 11 months using 3 different data sets, verifying responses against the respective documentation and assessing evolution risk. We found that there are: (1) inconsistencies in how these services behave; (2) evolution risk in the responses; and (3) a lack of clear communication that documents these risks and inconsistencies.

An Uber Technologies Inc. Volvo self-driving sports utility vehicles (SUV) sits on the road after a high-impact crash in Tempe, Arizona, U.S., on Friday, March 24, 2017. Uber Technologies Inc.'s self-driving cars were back on public roads Monday,... A spate of recent crashes involving Tesla and Uber are taking a toll on public confidence in self-driving cars. Two studies released this week, one by AAA and the other by Cargurus.com, The AAA survey was conducted in April, shortly after a self-driving Uber vehicle struck and killed Elaine Herzberg in Tempe, Ariz., and after a fatal accident involving a Tesla Model X operating in Autopilot mode in Mountain View, Calif. Uber halted its autonomous vehicle testing in Phoenix, Pittsburgh, San Francisco and Toronto after the death of Ms. Herzberg.