When we talk about AI, there seems to be a wide range of understanding as there are types of AI technologies available. But right now, let’s see if we can find some common ground on why we are, where we are. Let’s discuss some of the common constraints or roadblocks – whatever challenges that have prevented some of us from adopting artificial intelligence, and why we might not be there yet!
Sometimes we may simply have difficulty embracing changes to our regular routines. Or after embracing the plan to change, maybe we didn’t spend enough time on the requirements, the baseline or the expected benefits. Overall, sometimes we should be careful about what we ask for. Just like the guy in the Dilbert cartoon here:
Perhaps your business doesn’t have the amount of data required or the structure of data required to easily adopted artificial intelligence. You might not want to underestimate or trivialize these tasks, as they are very very important to the adoption success. In fact, you may have experienced that AI talent might be hard to find. Especially if your competitors or big business corporations have harvested a big portion of the so-called ‘experts’ in the AI resource marketplace. As we see in this graph, the demand for talent is rising fast, but the supply hasn’t moved up as quickly:
Even if you managed to acquire the so-called experts, maybe they aren’t the corporate fit that you were hoping for. Yet, if your ‘expert’ is OK and seems to settle in fine, are they playing well with others? Are you experiencing any issues outside of the expected forming, storming, norming, performing team dynamics? Can your new AI employee grow to be productive across your business enterprise, beyond your immediate project needs? These are very important things to consider. We have many years of successful IT delivery in custom software, integration, project management and more.
Does AI really have to be complex, difficult, unattainable? Of course not! We all know from our own personal experience that AI is very achievable. Just look at examples all around us. So Artificial Intelligence does look achievable, right? Let’s take a look at some of the most common and/or popular AI use-cases for business. Here are three that at least some of us can relate to:
In Supply Chain, for example, machine learning can use your data to help pinpoint the most influential factors to your supply networks’ success. Key factors are (1) influencing inventory levels, (2) supplier quality, (3) demand forecasting, (4) procure-to-pay, (5) order-to-cash, (6) production planning, (7) transportation management and more are becoming known for the first time.
In Human Resources we can use AI to:
- help customize the onboarding of new employees. Such as managing FAQ lists, allocating space, allocating new laptop, etc.
- Automate scheduling interviews, provide ongoing feedback to candidates and answer their questions in real time.
- Identify and remove unconscious language bias patterns to welcome diverse applicants.
- Recognize employees that may be heading for the exit door, aka ‘flight risk’.
All this contributes to keeping our employees happy and dynamically engaged.
Another great example, this time with ‘human in the loop’, rather than having AI totally replace the human in their job function: AI tools can train the CSR real-time with emotional intelligence tips. When the caller’s voice increases in pitch, volume and/or speed, the software assistant may recommend strategies to the CSR to reduce the caller’s stress to help the caller find their happy place. This would (1)increase call efficiency, (2)boost satisfaction, (3)gain unrivaled insight, and (4) continues improvement and learning will lead to better results. A happy caller makes a happy employee, and vice versa, leading to better client satisfaction, better client retention, and gain loyal customers.
In the banking and payments industry, there are many many opportunities for AI to automate tedious, time-consuming, error-prone human activities. Reduce costs, reduce errors, increase reliability, increase client satisfaction. Here is an infographic sourced from Forbes in 2017. Which they borrowed from Foresters Research.
As we started in 2005, located in Toronto, our team of experts has an extensive background in many disciplines of IT and is ready to compliment your existing IT team. We have partnered with many businesses in Toronto where our team has successfully delivered IT solutions. We also have been awarded Vendor of Record by the Ontario Government.
For hundreds of years, philosophers speculated that one-day human thought would be mimicked by machines. During WW2, Allan Turing and the “Bletchley Park” team created arguably the very first AI, used to defeat the German’s Enigma Machine. Soon after, in the 1950s, the word Artificial Intelligence was branded in Dartmouth college, and new A games such as checkers, chess, and GO started to become the first AI programs that attempted to solve complex problems. Not long after AI was introduced in robotics, reasoning and search engines, speech and language recognition and more. However, in the 1970s, an AI winter shut down all fields actively progressing in developing and maturing it. Yet, Japan started other attempts to bring back AI and progress the evolution of intelligence in modern hardware and software technologies.
One of the earlier examples of AI in business software was Microsoft’s Clippy, which was supposed to be helpful, though it was proven unuseful. In other case search engines, another AI tool such as Ask.com, Google, and Yahoo was starting to mature and help people to solve problems such as “how to disable Clippy and get on with our jobs.”
More recently AI is used in medical diagnosis to increase the quality of care. Financial trading is using it to increase unit-holder value based on real-time market factors. Also in Robotic Controls, AI is applied to increase the quality and consistency of surgery techniques on patients.
Remote sensors in cars, in industry, such as our own IoT or Industry 4.0, Tele-Sensor product from AnooshTech, to assist in predictive maintenance, rather than the old-fashion “preventative maintenance.” The Tesla car has a ton of AI, including very sophisticated power management. Another product, widely called ‘Auto-Tune’, helps music producers get the very best out of the talent that they have to work with.
As Dr. John Kelly III, IBM Executive Vice President said:
The success of cognitive computing will not be measured by any of Turing’s tests or a computer’s ability to mimic humans. It will be measured in more practical ways:
- return on investment
- new market opportunities
- diseases cured
- lives saved.
It’s important that your company or organization select the appropriate candidates for AI using value and risk as keys. Ensure that your infrastructure is ready, whether stand-alone or in the cloud. Data fabric definition and characterization are a very critical step. Start with a prototype, offline in a lab. When ready – i.e. when successful – then operationalize it. Ensure that you have the proper governance in place to safeguard the process. Drive the processes to help make the changes successful across the organization. Then evaluate the lessons learned to incorporate these into the next project and so on. And always remember “we are here to help you!” Let’s take a look at some examples of the application of AI in the market and industries.
We know that Google Lens and other AI are available for image processing. So let’s look at an application for it. Here is a camera capturing device that is used over four different garbage bins. Using AI and image processing and/or recognition this device is able to differentiate between the type of waste you have, can differentiate between parts of your waste and tell you which goes to which section – i.e. can, plastic, food, drinks, etc.- and post the information on the screen.
In addition, applications that are produced today to capitalized on human skills within Human Resources, are aiming to increase productivity and resource management. This will help with generating better results in your work environment such as the following:
- 91% reduction in time to shortlist
- 36 hours to reach your 1st qualified candidate
- 56% of candidates shortlisted are women
Within the current day, the financial industry should apply Machine Learning to use dynamic datasets, rather than relying on static (i.e. old) data. Current methods predict creditworthiness based on static information from loan applications and financial reports. This will them to identify current market trends and relevant news items that can affect a client’s ability to pay. Furthermore, in the investment prediction industry, fund management moving away from traditional market analysis methods. Using machine learning Major US banks are developing automated investment advisors, powered by machine learning technology. More importantly, machine learning can be used to help with fraud detection. This is great for analyzing high-volume financial account data. In this way, businesses protect their clients against fraudulent activity. This cost is almost $3 in recovery, for every $1 lost to fraud.
Anoosh tech can help your industry, to develop AI applications. We have decades of I.T. industry experience, skills in the latest technology – for AI, BigData, and IoT -, an expert team that can instantly increase your team capacity, an agile delivery method – fewer specs, faster results -, and more importantly we commit to your project success! We will review your system and provide AI feasibility advice.