AI: What Is It Good For? Plenty.
by Chris Nerney
If recent investment and acquisition activity are any indication, artificial intelligence (AI) – after years of seemingly slow progress and unfulfilled promise – is about to get its long-awaited moment in the enterprise spotlight.
Recent data from CB Insights shows the pace of acquisitions of AI startups by major companies has accelerated dramatically since 2011:
Among the busiest buyers have been Google, Apple, IBM, Intel, Microsoft, Salesforce, eBay, Facebook, and Yahoo. While some of these acquired AI technologies will be used for consumer products or services, many are targeting the enterprise.
- 2011 – 4
- 2012 – 7
- 2013 – 14
- 2014 – 33
- 2015 – 37
- 2016 – 42 (through Q3)
Given the amount of money these tech giants and others are betting on AI, it’s no surprise that AI startups continue to attract increasing amounts of venture capital. CB Insights reports that more than 550 AI startups raised a record $5 billion in global venture funding in 2016, while deal volume hit a five-year high of 658, more than four times the 160 deals in 2012.
The research firm also recently debuted its “AI 100,” a list of what it calls the most promising private companies applying AI algorithms across a number of industries. That list is as good a starting point as any to explore the current and potential use cases for AI in financial services, healthcare, and manufacturing, for starters. Here are a few that stand out:
New York-based Personetics has created an AI platform that helps banks better engage and respond to customers, offer personalized advice, and even take action on behalf of customers. The Personetics Cognitive Financial Services framework bundles built-in financial intelligence and conversational proficiency with advanced cognitive capabilities, resulting in an improved customer experience, greater online engagement, increased cross-selling, and lower customer support costs.
Silicon Valley startup Ayasdi’s machine intelligence platform is being used by financial services firms to improve fraud models, make risk forecasting faster and more accurate, better understand customers, and score better in financial stress tests. The platform is combined with a technique Ayasdi developed called Topological Data Analysis (TDA), based on the mathematical discipline of topology (the study of shapes and spatial relations). By applying TDA to massive, complex data sets, Ayasdi’s technology can reveal otherwise unnoticed critical patterns and relationships that can be turned into actionable data.
Healthcare and medical research
BenevolentAI is using AI and machine learning to accelerate medical and scientific research by enabling the mass analysis of huge amounts of highly fragmented scientific information, much of which is hard to find or access. Currently the London-based startup is focusing on the disease areas of inflammation, neurodegeneration (such as Parkinson’s, Alzheimer’s), orphan diseases (Amyotrophic Lateral Sclerosis, or ALS), and rare cancers. Future potential applications of BenevolentAI’s technology include women’s health, rare human diseases, veterinary medicine, and agri-tech.
iCarbonX, a startup based in Hong Kong, is building a massive “knowledge base” of health data from millions of people in order to provide individualized health analysis and a personalized health management system. The company recently formed a Digital Life Alliance (joined by seven other firms) with the goal of applying AI and sequencing technology to genetic, biological, and patient-generated data to “instantly detect meaningful signals about health, disease and aging” in an individual.
Manufacturing and industrial
Verdigris Industries, another Silicon Valley startup, is using AI to help enterprises with large facilities control operational and energy costs by deploying intelligent sensors that can instantly detect when equipment is consuming excessive energy, running erratically, or spiking. This allows IT and operations professionals to troubleshoot problems instantly and analyze energy usage remotely in real time. Verdigris also combines equipment performance data with temperature, occupancy, and utility data to understand what's "normal."
Sight Machine has developed a cloud-based manufacturing analytics platform that combines process and product data to drive immediate insights into productivity performance, process improvement, asset failure forecasting, and other manufacturing activities. The San Francisco-based startup’s platform is built to acquire and analyze data from myriad sources, including sensors, energy meters, images, quality systems, spreadsheets, logs, PCs, SQL, batch reports, and more.
There are many other AI startups targeting the sectors mentioned above as well as other industries, including automotive, retail, business intelligence, robotics, cybersecurity, commerce, and text analysis. This year should see a continuation of heavy investments in AI technologies that can help enterprises increase revenue, decrease costs, solve problems, and discover opportunities.
In which area of the enterprise do you think AI will have the greatest impact?