AI Today: Who Is Using It Right Now, and How
AI is a versatile tool, but how is it currently being used in business? Here we take a look at some of the implementations.
Artificial intelligence is all the rage in the enterprise these days. Stories abound about all the gee-whiz capabilities it will bring to our personal and professional lives.
But like any technology, there is usually a fair amount of hype before the reality sets in. So at this point, it is probably worth asking: Who is using AI right now, and how?
AI in Action
In a broad sense, says Information Age’s Nick Ismail, AI is already bringing five key capabilities to the enterprise:
- Voice/Image Recognition: Applications range from accurately transcribing meetings and sales calls to researching the impact of branding, logos and other visuals on the web.
- Data Analysis: Unstructured data in particular is very difficult to quantify. Using readily available tools, organizations are able to delve into the minutia of their operations, supply chains, customer relations and a wealth of other activities to gather intelligence that is both accurate and actionable.
- Language Translation: Convert one spoken language into another in real time, an increasingly important tool for multi-national corporations.
- Chatbots: Automate the customer experience with a friendly, responsive assistant that can intuitively direct inquires to the proper knowledge base.
- Predictive Analysis: Accurately forecast key data trends, such as cash flows, customer demand and pricing.
To see some of these capabilities in action, check out the new website for Peach Aviation which features an automated response system that provides multi-language support for customer inquiries. The system runs on the Desse AI agent provided by SCSK ServiceWare Corp., and can respond in all languages serviced by the airline: Japanese, English, traditional and simplified Chinese, Cantonese, Korean and Thai. As well, it uses data analysis to continuously monitor questions and answers to provide steadily improved quality. The company reports that out of 100,000 inquiries received in late December and early January, the system was able to provide automatic responses to 87 percent. (For more on AI in business, check out How Artificial Intelligence Will Revolutionize the Sales Industry.)
Yet another example of AI in action is a joint project by NBCUniversal and CognitiveScale to discern the key elements in a successful Super Bowl ad. The companies used CognitiveScale’s Cortex platform to analyze three years’ worth of game-day commercials and various client-engagement data to derive actionable insights linked to key video concepts, attributes and themes. For instance, the research showed that comedic effects work best with sales messages, while uplifting tones are more effective for branding.
While AI will not write and produce the perfect ad itself, NBCUniversal’s SVP of Corporate Analytics and Strategy Cameron Davies said it provides greater insight into what works and what doesn’t. “The CognitiveScale platform gives us the ability to consider new ad strategies for companies who want to ensure their ads will be successful when they invest in production and media buying,” he said.
CognitiveScale is also working with organizations in the financial, health care and retail industries by allowing video data to undergo the same analytics processes as voice, image and text.
AI is also turning into an effective crime-fighting tool, says Forbes’ Rebecca Sadwick. It turns out that one of the biggest hindrances to modern law enforcement is the bureaucratic inertia that exists in both public and private processes. AI helps overcome these hurdles, bringing much-needed clarity to highly organized criminal enterprises ranging from money laundering to human trafficking to terrorism.
One of the key ways AI helps solve crimes is by lowering the cost on private entities to oversee their transactions. Like any regulatory requirement, compliance is primarily a cost factor for organizations that are focused on profitability. Using third-party AI platforms specifically geared toward identifying suspicious data patterns, companies have not only lowered their costs but increased their chances of detecting nefarious activities. Prior to AI, it is estimated that nearly half of all financial crimes went unnoticed.
As well, banks and financial institutions that have deployed AI in this way actually help law-abiding citizens take part in fighting crime. Every time a legal transaction is processed, a learning algorithm is exposed to the normal patterns of money movement and is thus better equipped to identify transactions that break these patterns.
Technology is a two-way street, of course, so the same technology that is currently helping to fight crime can also be used to conduct it. With enough computing power, an intelligent system might be able to leverage the rising trend of micro-transactions by breaking up large transactions into numerous, perhaps millions, of smaller ones that are harder to detect and track. As well, quantum technology has yet to make its presence known in the criminal underworld (as far as we know, at least), which would open up an entirely new front in the war against cybercrime. (To learn more about quantum computing, see The Challenge of Quantum Computing.)
Clearly, we are in the earliest stages of AI development, so there will no doubt be numerous other ways in which it will affect mainstream enterprise processes as the market matures. Unlike earlier technologies, however, AI is expected to improve with age as it incorporates both human- and machine-generated data to forge a greater understanding of the environment it occupies and how best to navigate it.
And this is likely to be the most profound change of all: the end of lengthy development processes in which new features come out once a year (if that) and can only be implemented by taking infrastructure and data offline. In the future, digital systems will get better with age, all by themselves.
More from AltaML
- What are some of the foundational ways that career pros stand out in machine learning?
- What are some of the dangers of using machine learning impulsively without a business plan?
- Why is so much of machine learning behind the scenes - out of sight of the common user?
- What is 'precision and recall' in machine learning?
- Why are GPUs important for deep learning?
- What’s a simple way to describe bias and variance in machine learning?
- What are some of the main benefits of ensemble learning?
- Will machine learning make doctors obsolete?
- What's the difference between machine learning and data mining?
- How does a weighted or probablistic approach help AI to move beyond a purely rules-based or deterministic approach?
- Why is data visualization useful for machine learning algorithms?
- Why are some companies contemplating adding 'human feedback controls' to modern AI systems?
- What does 'connectionism' mean for business AI?