The Current State of AI Enterprise Automation

AI technology has become so ubiquitous that it's evolved into a very affordable enterprise solution. We explain how enterprises are incorporating automation into their day-to-day service offerings.

Once upon a time, AI technology was out of the reach of -- well, pretty much everyone. Governments were the first to fund research into the technology, followed closely by some of the world's largest organisations. Now, AI technology has become so ubiquitous that it's evolved into a very affordable enterprise solution. Not only that, but it’s no longer limited to IT professionals with a strong coding background. Today’s off-the-shelf AI tech is simple and sleek -- and is designed to be used by just about anyone in the business world. Which explains why it's becoming so widely adopted across the board by enterprises large and small (and every size in between). In fact, the question isn't any longer which enterprises make use of AI technology? But rather, which ones don't? And even then, you may struggle to find an answer to such a question.

In the next sections, I’ll get into how enterprises are incorporating AI automation into their day-to-day service offerings, with a particular focus on how it’s being used in customer service operations across the globe, as well as threats and solutions for the technology.

Need for Automation Strike Teams


While AI solutions are taking on the tedious and mundane tasks at the enterprise-level, new positions are cropping up in response to the new challenges AI technology brings with it. Automation strike teams, as they're being called, are the new crucial branch of your enterprise's IT department. These specialised units -- which we will see in increasing numbers starting in 2020 -- are being designed specifically to deal with situations where the AI goes awry or where it needs extra programming or upgrading. The most successful automation strike teams will be the driving force for automation at their enterprise, executing the strategy, planning, training, presenting and championing of the technology for the entire company.

AI technologies being adopted by enterprises


AI automation is being rolled out across enterprises in a variety of ways. The main technologies being implemented -- particularly in the customer service sphere -- are machine learning, deep learning, NLP/NLU, and computer vision. Let’s take a look at some of the ways these tools are being put to work in offices around the globe. Machine learning tends to be most successful when applied to large amounts of data. The most common way we’re currently seeing this form of technology being deployed as we enter 2020 is in the form of chatbots.

Look at Uber -- the company’s Michaelangelo chatbot helps Uber drivers respond to customer questions quickly and efficiently, by using NLP to pre-process and encode customer messages, then generating prediction scores for possible intent. The driver receives four suggested replies to a single customer question, based on the prediction scores, and he or she may then respond to the customer query with just one click.

US Supermarket Chain Harps food stores have been employing the technology to improve and optimise its chain-wide pricing and promotional strategy


Meanwhile, retail providers are making use of AI technology to automate time-consuming retail processes. Among such tasks is analysing transaction data to provide teams with insights to make better promotional and pricing decisions on products. US Supermarket Chain Harps food stores have been employing the technology to improve and optimise its chain-wide pricing and promotional strategy -- particularly focused on its weekly print circular (the company’s largest advertising expense). The system analysed years’ worth of transactional data and playing around with different actions to discover all the ways their customers make transactional decisions. With this data, the platform was able to offer Harps insights that helped the company “supercharge” its pricing and promotion decisions, resulting in a sales increase of just under 3 percent, and “significantly improved profits.”

Enterprise suites with AI


Off-the-shelf AI is, of course, the easiest (and least expensive) option for start-up and small-to-medium enterprises looking to incorporate AI technology solutions. But today’s AI providers are going beyond traditional software and offering more holistic platforms in the forms of suites and operating systems. One such example is Walmart’s use of HANA -- SAP’s cloud platform database server -- to process its huge volume of transaction records. Its main use to the US retail giant is its ability to access data in real-time (since it’s stored in RAM, rather than on the disk) for use with HANA’s applications and analytics, for faster decision making.

Walmart and other companies using the software are expected to benefit from infrastructure cost reduction and operational efficiency.


A Walmart factory manager, for example, maybe monitoring assembly line equipment using HANA. Meanwhile, data from a production slowdown could be collected and processed through HANA, which would then be used to determine if a new course of action is needed, such as an equipment service inspection. Walmart and other companies using the software are expected to benefit from infrastructure cost reduction and operational efficiency. In fact, SAP expects that its clients should see a 575 percent return on an investment after five years of using the software.

Concerns about data security


It hasn’t been all sunshine and rainbows when it comes to AI automation adoption in the enterprise. Concerns about data security are big, and mounting -- particularly in the wake of several large-scale data security breaches over the past few years. Capital One’s private server holding customer information was hacked in early 2019, exposing the personal information of over 100 million of its customers and applicants. The major weakness, in that case, was the fact that the information was hosted in the cloud. This particular cloud, you see, was hosted by Amazon Web Services. The hacker, it may not then surprise you, was a past employee of Amazon Web Services. Keep in mind that this is on the back of the 2017 Equifax data breach, which affected just under 150 million of its customers. The company only recently settled a lawsuit where it agreed to pay out up to $700 million to help compensate the victims of hackers who stole their personal data from Equifax’s servers.

The major weakness, in that case, was the fact that the information was hosted in the cloud.


All of this raised big concerns about cloud data security -- since storing big data in the cloud is likely the way forward when it comes to AI automation services. And because of this, some enterprises have been timid to adopt and implement any further AI technology until these concerns can be ironed out. According to an article in Towards Data Science magazine, “Until this serious, material, and valid cloud data security concerns are addressed, enterprise AI/ML adoption on the cloud will stall, relegated to small projects, delivered slowly, and unfortunately, providing limited value.”

Using AI to combat cyber attacks


Amazingly, though, AI is being used to help fight the very problem that it has helped escalate. Machine learning, it turns out, is really useful for helping to detect cyber threats before they happen. Computers can employ machine learning to do this by using and adapting algorithms based on data received (perhaps of past data breaches), learning from it, and then understanding what improvements an enterprise can undergo to help tighten up security.

Cambridge-based cybersecurity firm Darktrace, for example, uses AI technology to detect and halt new cyberattacks -- even those insider ones from former employees (since this seems to be something of a trend). Among its satisfied customers is Virgin Trains, whose IT Network Manager says of the tech:

“Darktrace’s cyber intelligence platform and Threat Visualizer interface provide us with absolute visibility into what is happening in real-time.”


With all these advancements in AI automation, it’s an exciting time to be an enterprise. If you want to be ahead of the game as we kick off 2020, there’s no time to rest. Get your hands on all these AI tools that will help you run circles around the competition by automating all your repetitive tasks without spending too much money, and take advantage of AI tools to protect your company against the threat of cybercrime.

Sources:

  • https://medium.com/syncedreview/how-uber-lyft-use-ai-to-improve-ride-experience-f7ffc722dedc
  • https://www.daisyintelligence.com/
  • https://emerj.com/ai-sector-overviews/ai-in-business-intelligence-applications/
  • https://www.darktrace.com/en/resources/ds-pov.pdf
  • https://towardsdatascience.com...
  • https://www.cnet.com/how-to/if...
  • https://www.nytimes.com/2019/07/29/business/capital-one-data-breach-hacked.html

 

Similar Articles

Discover the power of ultimate.ai

Register for a demo and one of our specialists will be in touch to show you ultimate.ai can transform your business.

Book a demo