As 2018 comes to a close, it's time to start thinking about the future of technology and what technological advances will be on the horizon in the coming year. Teradata's experts have compiled a list of their predictions for 2019 in the areas of artificial intelligence, data and analytics
, connected cars, security, blockchain, the cloud and much more. Take a look at their predictions and share your own with us on our social channels @Teradata
Todd Walter, Teradata's chief technologist
Currently, we’re in early days of an emerging market of connected cars. There are many things to be decided strategically and they will not all be settled in 2019 by any measure. However, as we move the needle towards an autonomous vehicle landscape, new partnerships, industries and challenges will emerge including:
There will be more partnerships and as a result there will be more mergers and acquisitions in the automotive and technology industries – a truly autonomous vehicle requires a variety of components to be linked together successfully. Currently, each vehicle manufacturer only has a few pieces.
Vehicle-as-a-service will continue to grow quickly – enabled by connected car technology. Uber, ZipCar, Maven are changing the definition of vehicle ownership. There will be a lot more headlines about how VaaS changes the business model of car manufacturers. The change will be slow but accelerating for the next couple years but the companies see the writing on the wall where there will be a lot less vehicle ownership and a lot more VaaS and they are trying to adjust their businesses to get ready for and hopefully take advantage of that.
The traditional vehicle manufacturers will have to move fast to stay ahead of the upstarts. Already traditional vehicle vendors like General Motors, Toyota and Ford are competing with technology vendors like Google, Uber, and Apple for the shares of the autonomous vehicle market. Vendors not perceived as keeping up on connected cars will be drastically punished on the market.
There will continue to be accidents with autonomous technologies. As a culture we focus on one fatal accident with the new technology rather than the hundreds of fatal accidents daily with human operated technology. This will lead to back pressure on the deployment and adoption of autonomous driving technologies. As a result, an uninformed cultural backlash against the technology is likely to occur. Conversely, positive outcomes are likely to take place including more stringent safety standards and a focus on transparency of information around incidents.
There will be a lot more focus on privacy and security of these new technologies. The information from the connected car is arguably even more sensitive than our credit card information. This information includes where we go, when do we go there, when are we home, where do we shop and work, where do our kids go to school and what locations do we go to at what time. There will be breaches of this personal information and bad things that happen as a result. There will be more of the takeover scenarios where an external bad actor can take over the technology. Takeover situations will result result in backlash and involvement of political and legal entities to begin to make laws and precedents. What can law enforcement access and discover to use for investigation purposes?
Atif Kureishy, Teradata's vice president global emerging practices
- Escalating Financial Crime - Synthetic Identify Fraud will continue to be a significant concern for a majority of companies involved in electronic payment. Many will have experimented using machine learning and several will have demonstrated improved performance over rules-based systems. Many of those companies will also have recognized the need to integrate information across different lines of business (e.g., HELOC, small business credit card, consumer credit) to enable early detection of fraud patterns using deep learning to better detect complex signals. Virtually all of companies will be challenged to deploy their new models due to operational and compliance issues dealing unless they can solve their model lifecycle and model risk management strategies.
- Auto Labeling – For supervised learning, large sets of human annotated data is needed to train a deep learning model that performs a particular task. A fundamental challenge that the Enterprise faces today in their AI journey is the creation of customized high-quality human annotated data. This process is slow, repetitive, may involve subject matter experts, and at times need to be redone. For enterprises, this is a significant upfront investment with a big risk and big costs. In 2019, we will see a trend towards AI powered tools that assist humans in the creation of high-quality annotated data through auto-labeling techniques. AI involvement at early stages of the journey will reduce cost, risk and help create efficiency, these will play a big role in fueling AI adoption at enterprises.
- Reinventing Retail - Brick-And-mortar retail businesses are turning their attention to AI to significantly improve customer experience, profitability and remain competitive. In 2019, we will see emergence of new data sources (surveillance cameras, on-the-shelf-cameras, robots) and AI models for inventory management, better customer retail experiences, targeted marketing, and adding new capabilities such as self-checkout. The key challenge, however, is to develop and scale AI operations to thousands of retail-stores that differ in planograms, camera models, and network infrastructure capabilities.
- Robust context-aware models – Today, machine learning models are trained with very narrow tasks in mind, with increasing specialization. The need for more generalized inference will lead to more models that perform joint estimation of disparate output types, replacing chains of specialized models that each perform a single task. For example, instead of creating separate models for detection, object tracking, motion forecasting, and motion planning, executed in sequence, a single model is created that performs all these tasks jointly. These models can benefit from internally reusing computations and sharing high-level features. In 2019, we will also see new system architectures that provide "context" to individual AI models, by arranging these models hierarchically and/or connecting them at different temporal or spatial scales.
- Data Minimization - Data valuation strategies will become increasing important. More data is better, right? Not always. Organizations are realizing that it's time to be more selective when it comes to data and more is not always better. In the future, we will see extra effort spent into discovering what is the true value of data to draft data minimization policies especially with the progress we see in making AI/Deep learning work with fewer data.
- Reinforcement Learning - To date there have been very few examples of applying reinforcement learning to enterprise problems like recommending the best offer to a customer or supply chain optimization. These examples, amongst others, can be complex problems for machine learning models because they may have multiple potential factors to optimize and involve a series of events leading to a decision or business outcome. Both of these characteristics make them well suited for reinforcement learning and the opportunity to drastically outperform current approaches. 2019 will have breakthroughs in the Enterprise for RL, with a focus on using off policy learning to overcome the challenge of not having a real-world environment to train the model like you would a robot in a lab.
- Industrial Inspection - Current smart camera solutions for manufacturing provide a generic software tool kit for product quality inspection that are intricate to tailor to specific requirements laid out by manufacturers. These black box smart camera software solutions limit manufacturers ability to combine features derived from image analytics with operational data that can improve early detection of quality issues. Furthermore, the software tool kits do not use state-of-the art analytical techniques. Consequently, such generic solutions do not cater to the wide variability in requirements resulting in low performances in detecting quality issues. In 2019, Manufacturers will seek a plug-and-play deployment of customized AI models that can provide high value at low cost/risk. Manufactures will need an Analytic Ops framework for ongoing monitoring, retraining and redeployment of models to continuously improve, which is again outside the purview of smart camera providers.
- Edge AI - With more tools being built to make deep learning models smaller and more energy efficient without sacrificing model performance, edge computing for AI will be more adapted and will improve human/AI interaction. This will lead to more robust applications in retail and manufacturing to drive better understanding of what is happening within the physical proximity of their enterprise.
- Model Risk Management - As the number of models deployed by enterprises grows, the need to manage model safety and model stability becomes increasingly clear. In Financial Services, Model Risk Management (MRM) is a well-known discipline to ensure the validity of models used in key processes such as credit underwriting. In 2019, other industries will realize the need to take a similar approach in order to realize the full value of their analytical models and data while managing risk exposure to their organization. Beyond model management, MRM provides data pipeline lineage, model governance, clear workflow for promoting models, reproducibility, stress testing, regulatory compliance, model performance monitoring, outlier detection and data monitoring.
- Humanized Digital Assistants - We have seen in 2018 big breakthroughs when it comes to AI assistants, synthesized human-like voice, and improved personalized dialog agents for the consumer. AI assistants will become more integrated and pervasive in many businesses with new opportunities for enterprise engagement in creative ways. Next-gen Business Intelligence will incorporate voice-based interfaces to support executive dashboarding and what-if scenarios.
Sri Raghavan, director, data science and advanced analytics product marketing at Teradata
- The rise of granular video analytics to aid in use cases in law enforcement, traffic & safety, and retail (among others)
- AIOps will be an oft heard construct. It will enhance existing Ops (IT and Development) capabilities with an algorithmic and automated approach towards decision making.
- The rise of Behavioral Biometrics: phones and other devices can recognize the user by the way they type, scroll or talk. This is to enhance device security and protect the owner from sensitive information leak.
- The rise of AI assistants everywhere will continue (e.g., cars, restaurants)
- Data storytellers will be given a greater hiring premium in companies that do analytics (which is mostly all).
- Most AI projects will mostly be in a “science experiment” stage as talent scalability to make them widespread is hard to achieve.
- Blockchain will slowly gain greater enterprise adoption but only after it distinguishes itself from the negative reputation of crypto currencies.
- A lot of advanced analytics will be powered by natural-language understanding and therefore increase the footprint of data driven decision making across many organizations
Chad Meley, Teradata's vice president of marketing
- The AI talent gap will continue to widen
- In a survey conducted by O’Reilly Media, the primary obstacle to successful adoption of deep learning is shortage of talent - recognized by respondents more than twice as much as other types of constraints.
- The talent landscape has huge implications, and is affected by a number of trends, most notably its uneven distribution. Currently, leaders like Google, Apple, and Amazon are absorbing a huge amount of the available talent, which goes on to serve their business models. While tech giant's battle each other for AI talent, it’s leaving many other companies -- both within tech and elsewhere -- on the outside looking in. The result is that industries such as energy, communications, manufacturing, and others are left behind despite their desperate need for innovation due to digitization.
- Huge salaries being commanded by AI practitioners is certain to grow the talent pool in time, however it won't close the ever-expanding AI talent gap anytime soon. To remain competitive, enterprises must develop a robust talent strategy, including hiring graduates, acquiring IP through mergers and acquisitions, or interacting with academia and partners who have done AI in practice.
- Object storage stops being synonymous with public clouds.
- Cloud object storage has been picking up steam in the enterprise by driving down costs associated with storing ever exploding data volumes. Further, cloud-based object storage allows developers to access and control data via APIs, ensuring integration with legacy and future applications. Cloud object storage also works across datacenters, allowing enterprises to scale and implement disaster recovery more effectively.
- Amazon’s S3 object storage currently has the largest market share, with Microsoft’s blob storage growing at a rapid pace. While object storage was born in the public cloud, look for private cloud offerings to realize the fastest growth for consuming object storage over the next few years, thereby debunking the current view that object storage is synonymous with public clouds. Onsite object storage deployments offer performance benefits because data doesn’t need to travel across less-than-consistent Internet connections. Unlike many public cloud solutions, on premises and as-a-service object storage offerings won’t charge monthly per GB, HTTP GET request, or data egress, helping to limit ‘hidden’ or unexpected costs.
Lawrence Latvala, Teradata industry consultant
- 2019 will see the broad acceptance of the new “financial multiverse,” in which disruptive platforms-as-a-service bleed across industry boundaries and force financial enterprises to build and participate in hybrid customer ecosystems. Open APIs will continue to change the nature of what it means to be a producer or a distributor, and pervasive data intelligence will drive hyper-personalized customer experience at decreasing cost. The common goal is to anchor the sentient customer enterprise.
Will Griffith, Teradata industry consultant
- In the coming year, it’s likely credit card delinquencies and charge-offs will spike upward, forcing card issuers to put aside more reserves for loan losses, tighten credit underwriting standards and ramp up collections capabilities. Analytics and artificial intelligence can help in all three areas.
Dave Rosal, Teradata industry consultant
- AI and machine learning will not replace accounting and finance professionals but will empower the professionals by reducing manual data entry and improving speed, accuracy and quality of analytics which they provide to the business. In 2019, the digital transformation of finance will highlight the woefully under-documented current business processes resulting in significant delays in the implementation of major finance projects.
Eric Gookins, Teradata industry consultant
- 2019 will see the introduction of analytics and data leveraged in identity-based consumer and merchant products to create value for payment networks and providers.
Steven Warner, Teradata industry consultant
In 2019, retail banks will continue their migration in decisioning approaches from heuristic (SME-based rules) to statistic (predictive stats) to algorithmic (deep learning), to become more contextually-relevant to customers’ interactions with them, at lower cost.
Ted Vandenberg, Teradata industry consutant
Companies will mobilize to replace internal systems that help data scientists build and deploy models faster. That means databases, development servers and production scoring servers along with their interfaces. In most companies, the technology was not built for this purpose and cannot perform at any kind of scale; some call this the end state intelligent cyber physical systems (the term was coined in 2006 by Felen Gill of the National Science Foundation).
Michael D Halula, Teradata industry consultant
- Frictionless Experience – Retailers are looking to integrate their physical and digital channels to create a single shopping experience. The challenge is to incorporate a way to integrate their interactions as the customer purchase in the store, buys additional affinity products online, selects ship to home and to track their purchase through fulfillment. Retailers are investing billions in evolving the physical shopping experience. For example, Walmart.com redesigned their online site in 2018, Target is redesigning their physical stores and Amazon launched Amazon Go and acquired Wholefoods.
- AI Adoption – One of the most singular requests I get is how a retailer can adopt AI in their business processes ranging from supply chain, fulfillment, instore stock replenishment, to automated personalized communications through the purchasing process. Walmart, Lowes and other retailers are testing robotics instore to improve shelve replenishment/ stocking to automating the pick and pack processes.
- Last Mile/ Logistics – Every retailer is focused on the fulltime process. All companies are investing, partnering or through acquisitions to enhance the cost effectiveness of fulfillment. Retailers are using BOPIS towers/ lockers (Walmart, Amazon), to partnerships with other retailers (Amazon & Kohl’s), to acquisitions (Walmart & Parcel). The emphasis has spark initiates by retailers to improve their supply chains (Walmart invested $11B, Target invested in their new SC Flow centers $7B) to improve the speed to deliver while reduce the cost of goods sold.
- Physical Experience – Walmart, Target, Macy’s to name a few are all investing in the store experience. Macy’s Market@Macy’s pop-up store concept & their off brand Backstage stores will change the way customer’s shop with Macy’s. Target investing in a complete physical store makeover (1,000 stores across the country by the end of 2020) will provide a new, exciting shopping experience.
Jay Irwin, director of Teradata’s Center for Enterprise Security
- Data breaches will not see significant reduction even though emergence of a new mass of security and especially privacy law is trending globally. This is partially because statistics and our own experience show that customer adoption rates into programmed efforts to comply with these new reflects a “wait and see” attitude. The one that may change all that is California’s Consumer Privacy Act or CCPA, effective January 1, 2020, which is local, yet international, and definitely predictive of 49 more variations of such a law in the US over the coming 8-10 years. The best news that may mitigate my prediction some is that encryption of data at rest is on the uptick as organizations understand the protection value of lowering risk which being enabled to analyze all their data, and as technology providers release new ways to make the job easier and cheaper.
Mikael Bisgaard-Bohr, Vice President Business Development
- 2019 will be the year where “we” (vendors, consultants, customers, analysts etc.) will go back to the future in the sense that we will all come to understand that most of the value derived in our space comes from solving “old” use cases (churn, fraud, risk, cost etc) better through the use of more data, and not from addressing new and more esoteric use cases. I am sensing that in a lot of organizations the focus is moving away from new and shiny use cases and moving towards use cases that organizations have tried to solve for years. The difference is that we are now able to provide more solutions than ever before, thanks to better (more, wider and deeper) data sets and at scale thanks to advances in technology.
- 2019 will be the year when “data plumbing” – the tedious work of acquiring, managing and integrating data – will be recognized as the key ingredient for success if you want to be a digital company. Without good data feeding, organizations will never get any value from AI and the challenge is how to manage and serve data at scale. The organizations that can successfully do acquire, manage and integrate the right data with AI technology, will be the winners of 2019.
Brian Wood, Teradata's director of cloud marketing
- As-a-service Ascendance: The rapid proliferation of point solutions in the cloud for virtually every facet of the analytic ecosystem is fine for small, experimental, or clean-sheet projects where an individual can piece together and manage the components on his or her own, but entirely unwieldy and unsustainable for analytics at scale across distributed data sets. In 2019, we will see enterprises start to recognize the limitations of point products and instead embrace "full-stack" as-a-service solutions that integrate with, and enable, hybrid analytic ecosystems so that business analysts and data scientists can focus on creating rich, differentiated insights from disparate data sets to drive true competitive advantage.