In its recently published report onTechnology, Media and Telecommunications Predictions, Deloitte predicts that over 300 million smartphones will have on-board neural network machine learning capability in 2017, allowing devices to perform machine learning tasks even when offline. This will enhance applications such as indoor navigation, augmented reality, speech recognition, and language translation. Such capabilities are also likely to be found in tens of millions drones, tablets, cars, virtual or augmented reality devices, medical tools, etc. Deloitte also expects that DDoS attacks, including those involving compromised IoT devices, will be more intense and frequent in 2017. The company, however, notes that embedded machine learning capabilities have the potential to protect devices (though, for example, detecting malware or suspicious or anomalous behavior) and 'might help turn the tide against the growing wave of cyber-attacks'.