Cybersecurity was all over the news in 2016 – whether it was email breaches that compromised the Democrat campaign for the elections, or revelations towards the end of the year that planes were vulnerable to hacking through in-flight entertainment systems. The British government boasted that it had the capabilities to launch cybersecurity offensives and was committing a huge chunk of its budget to developing these further. Yahoo suffered from an attack that potentially gained access to 1 billion accounts, the largest known breach of all time. Vendors, hackers, banks, businesses, countries and shadowy state actors all seem locked in a perpetual game of cat and mouse – and highly sophisticated and organised malicious attackers seem to have the upper hand. According to the experts, here are some of the cybersecurity nightmares organisations will have to wrangle with in 2017.

No statistical algorithm can be the master of all machine learning application domains. That's because the domain knowledge encoded in that algorithm is specific to the analytical challenge for which it was constructed. If you try to apply that same algorithm to a data source that differs in some way, large or small, from the original domain's training data, its predictive power may fall flat. That said, a new application domain may have so much in common with prior applications that data scientists can't be blamed for trying to reuse hard-won knowledge from prior models. This is a well-established but fast-evolving frontier of data science known as "transfer learning" (but goes by other names such as knowledge transfer, inductive transfer, and meta learning).

And so the insurrection is beginning. Last week Japanese insurance Fukoku Mutual Life Insurance announced that it was going to be replacing 30 staff with an artificial intelligence that would be calculating payouts (although, it noted, with human oversight still making final approvals). The technology would improve productivity by 30% and the firm expected to save some 140m Yen a year (around £1m) after the 200m Yen investment. Now I'm sure that this implementation of IBM's Watson technology (remember: Watson was the man who predicted in 1943 a global market for maybe five computers) will be very whizzy. But excuse me whilst I contend that Fukoku's PR make this AI sound like every IT business case I've ever seen: cost savings through headcount reduction blah blah, productivity gains blah blah.

"Jia Jia" can hold a simple conversation and make specific facial expressions when asked, and her creator believes the eerily life-like robot heralds a future of cyborg labour in China. Billed as China's first human-like robot, Jia Jia was first trotted out last year by a team of engineers at the University of Science and Technology of China. Team leader Chen Xiaoping sounded like a proud father as he and his prototype appeared Monday at an economic conference organised by banking giant UBS in Shanghai's futuristic financial centre. Chen predicted that perhaps within a decade artificially intelligent (AI) robots like Jia Jia will begin performing a range of menial tasks in Chinese restaurants, nursing homes, hospitals and households. "In 5-10 years there will be a lot of applications for robots in China," Chen said.

The most potent and under-appreciated corporate function today is Corporate Communications. Their continued reliance on a pre-digital, centrally-governed model that delivers tightly-edited messages through pre-approved channels does not help. This approach is colliding with the new digital ways in which people search, find, consume and share news (including corporate information, real or fake). It's time for Chief Communications Officers to update their standard operating model to calibrate for the proliferation of decentralized digital networks as well as conversational user interfaces that are reshaping public and private modes of communication. The purpose of Corporate Communications is to be the primary source of truth for stakeholders, a role crucial for executives and society alike.

Maroon 5 keeps popping up on my Pandora stations, so artificial intelligence (AI) and machine learning still have a ways to go. Even if AI can beat us at Go. But, wow, that aside, the technologies showcased at the 2017 Consumer Electronics Show (#CES2017, actually the 50th annual, sponsored by the Consumer Technology Association [CTA]), from countless robots to Hyundai exoskeletons, are incredible. Voice recognition, virtual and augmented reality, smart home technologies and drones are everywhere. AI and machine learning comprise major threads. Which raises the question: Will all that AI be Democrat or Republican?

The Managing Partner of Intergroup Partners AG, Montserrat Corominas, has recently been published in Inc Magazine. This is her article "Artificial Intelligence and Virtual Reality are about to transform business. Here's how to be prepared" KEY OBSERVATIONS Most business leaders are not quite sure yet how AI and VR are relevant to their companies and what is in it for them. To my understanding, these are only contemporary descriptions of state of the art technologies. The most important topic to discuss for boards, founders and management teams in regards to AI/VR is the inherent new way of communication through digitization.

In recent years, computers have gotten remarkably good at recognizing speech and images: Think of the dictation software on most cellphones, or the algorithms that automatically identify people in photos posted to Facebook. But recognition of natural sounds -- such as crowds cheering or waves crashing -- has lagged behind. That's because most automated recognition systems, whether they process audio or visual information, are the result of machine learning, in which computers search for patterns in huge compendia of training data. Usually, the training data has to be first annotated by hand, which is prohibitively expensive for all but the highest-demand applications. Sound recognition may be catching up, however, thanks to researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL).

A scarf designed to filter out harmful elements in city air. A breast pump that fits into a bra and keeps track of pumping volume. A drone that can dive into water and help anglers catch a big one. CES 2017, the consumer tech event held in Las Vegas this week, featured vendors with automated baubles, humanoid robots and "smart" everything -- a dishwasher, hairbrush and lawnmower, to name a few. But it's not so much the gadgets as their underlying technology that will make CIOs -- who seek out new tools for business, not the home or yard -- stop, look and listen.

Looking at chatbot builders, I see a lot of technologists. People who are interested in AI and conversational UI, NLP/NLU and machine learning. I am by no means an expert in any of these fields. The little I do know though is around the idea of syntax and the difficulties of NLP in parsing language. Last year google released SyntaxNet to help with NLU.