Executive Insights on Artificial Intelligence and All of Its Variants
Executive Insights on Artificial Intelligence and All of Its Variants
To gather insights on the state of artificial intelligence, machine learning, deep learning, NLP, predictive analytics, and neural networks, we spoke with 22 executives who are familiar with AI.
Join the DZone community and get the full member experience.Join For Free
Bias comes in a variety of forms, all of them potentially damaging to the efficacy of your ML algorithm. Read how Alegion's Chief Data Scientist discusses the source of most headlines about AI failures here.
To gather insights on the state of artificial intelligence (AI) and all its variants — machine learning (ML), deep learning (DL), natural language processing (NLP), predictive analytics, and neural networks — we spoke with 22 executives who are familiar with AI.
- Gaurav Banga, CEO, and Dr. Vinay Sridhara, CTO, Balbix
- Abhinav Sharma, Digital Servicing Group Lead, Barclaycard US
- Pedro Arellano, VP Product Strategy, Birst
- Matt Jackson, VP and National General Manager, BlueMetal
- Mark Hammond, CEO, Bonsai
- Ashok Reddy, General Manager, Mainframe, CA Technologies
- Sundeep Sanghavi, Co-founder and CEO, DataRPM, a Progress Company
- Eli David, Co-Founder and Chief Technology Officer, Deep Instinct
- Ali Din, GM and CMO, and Mark Millar, Director of Research and Development, dinCloud
- Sastry Malladi, CTO, FogHorn Systems
- Flavio Villanustre, VP Technology LexisNexis Risk Solutions, HPCC Systems
- Rob High, CTO Watson, IBM
- Jan Van Hoecke, CTO, iManage
- Eldar Sadikov, CEO and Co-founder, Jetlore
- Amit Vij, CEO and Co-Founder, Kinetica
- Ted Dunning, PhD., Chief Application Architect, MapR
- Bob Friday, CTO and Co-founder, Mist
- Jeff Aaron, VP of Marketing, Mist
- Sri Ramanathan, Group VP AI Bots and Mobile, Oracle
- Scott Parker, Senior Product Marketing Manager, Sinequa
- Michael O’Connell, Chief Analytics Officer, TIBCO
The key to having a successful AI business strategy is to know what business problem you are trying to solve. Having the necessary data, having the right tools, and having the wherewithal to keep your models up-to-date are important once you’ve identified specifically what you want to accomplish.
Start by looking at your most tedious and time-consuming processes. Identify where you have the greatest risk exposure for failing to fulfill compliance issues as well as the most valuable assets you want to protect.
Once you’ve identified the business problem you are trying to solve, you can begin to determine the data, tools, and skillsets you will need. The right tool, technology, and type of AI depend on what you are trying to accomplish.
Companies benefit from AI by making smarter, more informed decisions, in any industry, by collecting, measuring, and analyzing data to prevent fraud, reduce risk, improve productivity and efficiency, accelerate time to market and mean time to resolution, and improve accuracy and customer experience (CX).
Unlike before, companies can now afford the time and money to look at the data to make an informed decision. You cannot do this unless you have a culture to collect, measure, and value data. Achieving this data focus is a huge benefit event without AI since a lot of businesses will continue to operate on gut feel rather than data. They view data as a threat versus an opportunity, and ultimately these businesses will not survive.
Employee engagement and CX can be improved in every vertical industry, and every piece of software can benefit. AI can replicate day-to-day processes with a greater level of accuracy than any human, without downtime. This will have a significant impact on the productivity, efficiency, margins, and the risk profile of every company pushing savings and revenue gains to the bottom line.
Companies will be able to get to market faster and cheaper, with greater customer satisfaction and retention.
The biggest change in AI in the near-term has been the fact that people and companies are beginning to use it, and all of its variations, to solve real business problems. Tools and libraries have improved. The cloud is enabling companies to handle data at scale necessary for AI, machine learning (ML), deep learning (DL), natural language processing (NLP), and recurring neural networks. In addition, more investment is being made in AI initiatives as companies see the dramatic impact it can have on the bottom line.
We’ve moved from machine learning to deep learning. We see more AI/ML libraries that are more mature and scalable. We see larger neural networks that are deeper and able to handle more data, resulting in more knowledge and greater accuracy.
Today the cloud is a commodity, and it’s possible that this will happen to AI as well, except faster, as consumers adopt autonomous cars and manufacturers put hundreds of millions of dollars on their bottom lines. AI improves quality of life for individuals, making things simpler and easier while improving the quality of life of workers and making companies significantly more profitable.
The technical solutions mentioned most frequently with AI initiatives are TensorFlow, Python, Spark, and Google.ai. Spark, Python, and R are mentioned most frequently as the languages being used to perform data science while Google, IBM Watson, and Microsoft Azure are providing plenty of tools for developers to work on AI projects via API access.
The real-world problems being solved with AI are diverse and wide-reaching, with the most frequently mentioned verticals being finance, manufacturing, logistics, retail, and oil and gas. The most frequently mentioned solutions were cybersecurity, fraud prevention, efficiency improvement, and CX.
AI helps show what’s secure, what’s not, and every attack vector. It identifies security gaps, automatically freeing up security operations to focus on more strategic issues while making security simpler and more effective.
A large-scale manufacturer milling aircraft parts used to take days to make the parts with frequent manual recalibrations of the machine. Intelligent behavior has increased efficiency of the operators, reduced time to mill apart, and reduced the deviations in parts. AI automation provides greater support for the operators and adds significant value to the bottom line.
The most common issues preventing companies from realizing the benefits of AI are a lack of understanding, expertise, trust, or data.
There’s fear of emerging technology and lack of vision. Companies don’t know where to start, they are not able to see how AI can improve their business. They need to start with a simple proof of concept with measurable and actionable results.
Tremendous skillsets are required. There’s a shortage of talent and massive competition for those with the skills. Most companies are struggling to get the expertise they need for the application of deep learning. Companies also have a hard time wrangling all of their data, which may be stored in multiple places.
Brownfield equipment owners can be very uncomfortable with anyone interfering with their very expensive and precise equipment. Luckily, you don’t need to touch their equipment to execute a proof of concept. Once a client becomes comfortable with what AI can accomplish, they are open to automation. Once end users see the data is more accurate than their experience, they begin to trust the data and trust AI to improve the efficiency and reliability of their equipment.
The greatest opportunities for the implementation of AI are ubiquitous — it’s just a matter of which industries adopt and implement it the quickest. All prospects have the same level of opportunity with AI. How can businesses identify jobs that require a lot of repetitive work and start automating them?
There are opportunities in every industry. We see the greatest opportunities in financial services, healthcare, and manufacturing. In manufacturing and industrial IoT, ML is used to predict failures so companies can take action before the failure, reduce downtime, and improve efficiency.
There are several well-known fraud controls. Companies can know what’s on the network, who’s on the network, what devices they are accessing the network with, what apps they are running, whether or not those devices are secure and have the latest security updates and patches. This is very complex in a large organization, and AI can handle these challenges quickly and easily.
The greatest concerns about AI today are the hype and issues around privacy and security. The hype has created unrealistic expectations. Most of the technology is still green. People are getting too excited. There’s a real possibility that vendors may lose credibility due to unrealistic expectations. Some vendors latch on to “hot” terms and make it difficult for potential clients to distinguish between what’s hype and what’s real.
As AI grows in acceptance, privacy and data security come into play, since companies like Amazon and Google hoard data. Who decides the rules that apply to a car when it’s approaching a pedestrian? We’re not spending enough time thinking about the legal implications for the consumer regarding cyber attacks and the security of personally identifiable information (PII). We’ll likely see more malware families and variants that are based on AI tools and capabilities.
To be proficient in AI technologies, developers need to know math. They should be willing and able to look at the data, understand it, and be suspicious of it. You need to know math, algebra, statistics, and calculus for algorithms; however, the skill level required is falling as more tools become available. Depending on the areas in which you want to specialize, there are plenty of open-source community tools, and the theoretical basics are available on sites like Coursera.
Opinions expressed by DZone contributors are their own.