Artificial Intelligence (AI) has been around for a while, yet new developments are being made every day. 2017 saw investors pump over $15.2 billion in funding into AI startups across multiple industries. It was a significant 141% spike in funding from 2016.
The ability to have machines process information with capabilities that both match and extend the ability of the human brain will have major repercussions in our society. AI is set to power the fourth industrial revolution- infiltrating into core aspects of everyday life.
So what can we expect in 2018? Both technological innovation and a wider adoption of AI will be key trends. As the data-sphere expands, concerns on privacy are becoming more apparent. Here are 7 trends to expect for AI in 2018:
1. Advanced Content Creation
Advancements made in Natural Language Processing (NLP) and Natural Language Generation (NLG) allow companies to churn out readable, engaging content in a number of seconds. AI content generation is already used in social media, financial reporting and marketing. It will only gain greater acceptance. We even use it in our own product- The Story.
2. A Robotic Workforce
Labour intensive work in assembly lines in factories are set to be replaced by AI programmed robots. In Japan, more than 80% of care-work for the elderly will be performed by robots. Tianyuan Garments Company in Little Rock, Arkansas will use sewing machine robots to manufacture T-shirts.
It is not just about robots in factories, AI is transforming the way we work. As more companies realise the benefits of AI- a higher number of tasks will be able to be automated using AI tools. Low-skilled tasks that take time away from high-skilled workers will be the starting point. We are already doing this by knocking hours off manager’s clocks using our retail data integration software to combine more systems into one giant data lake.
This will not benefit all industries. Professions like law bill clients on hours. AI has a huge potential to reduce time and improve efficiency of legal work. As it becomes cheaper and more widely adopted, law firms will have to restructure how they charge their clients.
Other industries, like health and medicine, will be affected. Drug discovery and medical assistance will be revolutionised by AI systems. AI will even extend into the creative sphere generating music, images and visual arts used in commercial products.
3. Better and Cheaper Hardware
New developments and cheaper hardware enable better data collection and faster deep learning. New processes with new software frameworks and interconnects are expected to emerge from established hardware companies, cloud providers and startups in the West and in China. These are designed for the sole purpose of companies speeding up their deep learning experiments.
New startups are entering the industry- reducing the price of specialised sensors, like LIDAR, used to collect the data used behind deep learning. Alternative sensors and new methods, including high-volume, low-resolution data from edge devices and sensor fusion will drive costs down significantly.
This will make AI more affordable and therefore expand its reach across industries.
4. More Voice Assistants
The Consumer Electronics Show highlighted the growth of voice enabled computing in 2018. Barely any Internet of Things (IoT) device was without integration into the Amazon Echo or Google Home.
The new Samsung Galaxy S8 included the launch of Bixby – a smart voice assistant much like Siri. Samsung aims to connect all its devices to Bixby by 2020.
5. AI will Expand into Offline Spaces
AI is mostly used as an online solution, yet companies like Amazon are looking at using offline data as well. They will collect data via sensors and actuators in malls and stores to track consumer’s movements and shopping habits. Amazon Go grocery stores already are taking advantage of the technology.
6. A Greater Concern for Privacy Safety and Security of Data
The launch of new GDPR regulations in Europe are an example of a heightened awareness surrounding data security, privacy and safety. Companies like Apple will work to develop or improve privacy-preserving machine learning products. Information will now not be shared to third party entities unless given clear permission from the information’s owner to do so.
AI will expand its reach into cyber security. One enterprise’s AI can spot Trojan apps on worker’s mobile devices and in app stores. This allows them to find “fake” apps and do additional analysis using traditional threat modelling to determine if these apps are real threats. Cyber threats can thus be vanquished.
7. AI will Become a DIY Project
Open source software libraries, hundreds of API and SDKs, along with assembly kits from Amazon and Google will allow more people to “skill up” and take a shot at building their own AI systems. Google has launched an AIY (Artificial Intelligence Yourself) project, complete with a voice recognition kit for Raspberry Pii. Users will be able to choose a voice for their personal assistants. As many industries will be placing a high price on AI skills, it is not such a bad idea after all.