Heard on the Street – 9/18/2023

Print Friendly, PDF & Email

Welcome to insideBIGDATA’s “Heard on the Street” round-up column! In this regular feature, we highlight thought-leadership commentaries from members of the big data ecosystem. Each edition covers the trends of the day with compelling perspectives that can provide important insights to give you a competitive advantage in the marketplace. We invite submissions with a focus on our favored technology topics areas: big data, data science, machine learning, AI and deep learning. Enjoy!

How AI will transform client service in Big Law and beyond. Commentary by Paul Giedraitis, CEO/Founder of Orgaimi 

“This year has been a breakthrough year for artificial intelligence, and the legal industry is now paying close attention to how they can best utilize technological solutions, including AI, to optimize their business and services. Looking to AI-driven solutions – specifically those producing predictive client intelligence – will be crucial if law firms want to solidify first-mover advantage status across the industry, gain a competitive advantage, as well as future-proof their organizations against economic volatility in the years ahead. 

The emergence of client intelligence data marks a pivotal shift in the modus operandi of law firms because it has redefined how they can function, measure impact, and engage with their clientele. In this context, client intelligence data refers to the collection, analysis, and interpretation of information related to client behavior, preferences, interactions, and feedback – actionable data of which law firms have no shortage. Law firms are sitting on enormous volumes of this data, which if properly integrated and consolidated, can help proactively address customer concerns, forecast demand/workload requirements, and ensure team composition that is not only at the right skill level, but diverse and inclusive, to address the client’s unique needs and concerns.

For Knowledge Management and Marketing teams at law firms, client intelligence data can  be a game changer in achieving business objectives, especially when it comes to executing laser-focused marketing campaigns, and conferring unparalleled personalization upon the entire client journey. With the right integrated AI platforms, law firms are employing data-driven decision-making to  generate client insights for content creation, analzye communication patterns to determine the most effective ways to engage with different clients, and predict client churn and potential legal risks specific to each client. 

It will be up to law firms, however, to adopt and incorporate the right data analytics technologies and data governance procedures to take full advantage of the growth opportunities their client intelligence data provides. 

Ultimately, firms that can action this higher quality client intelligence will foster a deeper sense of loyalty and engagement from clients, and lead to longer lasting and higher value client relationships. As the legal industry races to catch up to digital transformation, client intelligence is an essential dataset that directly supports efforts by law firms to future-proof their organizations, building a bridge between a firm’s aspirations and the evolving demands of a discerning clientele, as well as firmly establish the competitive edge they need to keep up with the tech-savvy Joneses.” 

AI healthcare solutions can transform global health equity. Commentary by Sahir Ali, Founder of Modi Ventures

“Artificial Intelligence (AI) is poised to usher in a transformative era for healthcare in the developing world, bridging gaps in global health equity and positively impacting the lives of billions. While advanced medical resources are often concentrated in urban centers, millions across countries like Africa, India, and East Asia lack access to sufficient healthcare professionals. This glaring disparity results in high costs, lost wages, and limited medical care for those residing in remote areas. However, the convergence of AI and digital health along with telemedicine holds immense promise in addressing these challenges.

For many individuals, seeking medical assistance involves a day-long endeavor, often requiring them to travel long distances. This scenario not only incurs excessive costs but also leads to significant wage losses, perpetuating a cycle of economic vulnerability. Enter AI and telemedicine, game-changing solutions capable of triaging the influx of healthcare cases and offering remote medical consultations. In many regions, scarcity of medical professionals is a pressing concern as well. Even rural hospitals equipped with imaging tools often lack the necessary personnel, oftentimes a handful of radiologists and pathologists available for millions of people. This deficit, however, is being combated by cutting-edge AI and medical imaging technologies that introduce revolutionary teleradiology and digital pathology solutions.  

In many areas of the world, governments are now placing telemedicine booths in villages that allow people to access physicians in a remote setting. In these areas, new technology such as AI-based medical imaging and digital auscultation (remotely listening to heart and lung sounds using machine learning), are massive improvements to healthcare. Infection is a fundamental reason for sickness in these areas and is very difficult to diagnose without auscultation of the heart and lungs of the patient. Many countries are experimenting with drone-delivered pharmaceuticals. These new technologies super-charge the effect of individual physicians, allowing them to address a much wider number of patients from a much larger geographic area.

Global health equity is about access to quality healthcare. AI and new medical technology opens up massive potential for improving the lives of people throughout the developing world.”

2.6 million DuoLingo users’ data was scraped and is being sold online. Commentary by Stuart Well, CTO at Jumio

“While the recent cybersecurity focus has centered around generative AI and protecting against new fraud techniques, this situation is a reminder that fraudsters still revert to tried-and-true methods of data theft. The 2.6 million users whose usernames and email addresses were scraped, now face increased phishing attacks as their data is being sold on hacking forums. If victims fall for the phishing scams, they are further exposed to additional attacks that can steal even more vital data like personally identifiable information (PII), passwords, and financial information. 

Cybercriminals will continue to leverage known methods of attack and develop new and innovative ways to steal data, meaning organizations must prepare for everything from the new to the old. In order to better equip themselves against all forms of fraud, organizations must build a strong foundation starting with user verification and authentication. It’s essential for organizations to implement digital identity verification to prevent attacks by cross-referencing the biometric features of an onboarded user with those of the cybercriminal attempting to breach the company. By establishing every user’s true identity, businesses can ensure the user accessing or using an account is authorized and not a fraudster, keeping user data out of fraudsters’ hands.”

GenAI’s dirty little secret — sloppy practices mean mistakes, errors and omissions. Ansh Kanwar, SVP Technology, Reltio

“AI/ML technology has quickly evolved to become a critical tool for companies in nearly every industry. It offers enormous potential for solving long-standing problems and dramatically accelerating business growth. Capitalizing onAI/ML’s potential will split companies into two categories:  those that learn to harness AI to drive their business effectively and those that disappear. 

AI is only as strong as its data foundation, however, so companies that lay the data groundwork now have a much better chance to reap AI rewards for years to come. AI tools are quickly becoming required for relevancy. Still, leaders are finding – much like cloud and digital transformation – that buying and implementing the latest tech and tools alone will not revolutionize a business. Poor data quality stands as the biggest roadblock to successful AI adoption and business transformation. That’s why every company must align on a solid data strategy before launching an AI strategy. Even advanced AI systems can falter without clean, unified, and relevant data. Combining bad data with AI tools only sets institutions up for failure. The adage, “garbage in = garbage out,” is a hard reality with AI tools. Unchecked, AI could worsen matters, leading to a runaway train of problems such as elevated security and compliance risks, bias in models, and inaccurate decisions. 

Once companies have a clear data strategy, it’s important to note that integrating AI is not a one-and-done event. The need for accountability grows as AI systems become more integrated and complex. Business decisions, insights, customer interactions, and innovations hinge on the trustworthiness of the data feeding the AI models. Every AI model must be accountable to a source. The criticality of increasing trust in data—in all corners of an organization – cannot be overemphasized. Trust is indeed the bedrock of any system, and in the context of AI/ML and enterprise data management, trust implies that data is accurate, reliable, consistent, and compliant.”

AI is great for developers, but can DevSecOps keep up? Commentary by Derek Holt, CEO of Digital.ai

“AI concepts and tools like AI-code assist, generative AI, and large language models help developers create and optimize code, helping to dramatically increase developer productivity. AI also helps organizations facilitate a faster developer onboarding process, and helps teams synthesize information almost instantly. However, with all these improvements developers receive with the help of AI, the question becomes, can DevSecOps processes, teams, and tools keep up? The risks are not small – bottlenecks can develop in release pipelines, as DevOps and security practices struggle to keep pace with increased developer productivity driven by AI; security and quality concerns may come about, as AI-code assist tools might misinterpret what code is being asked for, and developers could use code leveraged from unknown sources.  Organizations may become so concerned with data privacy, IP and copyright issues of AI-based products, that they shut down the use of AI altogether, losing out on a once in a generation productivity-boost opportunity.

In the age of AI, businesses need to support their developer productivity gains and innovations while overcoming these challenges. Here are three core concepts a business can leverage to best utilize AI. ​First, by focusing on the fact that developers are becoming more productive and faster at generating code, teams can shift traditional software delivery workflows to also leverage AI capabilities to further automate software delivery. AI-augmented concepts like predictive planning, automated testing, predictive threat analytics, or automated deployment processes can be implemented so DevOps and security teams can handle the increase in the production of code. 

It’s not sufficient to deliver code into production faster through automation; the code needs to be delivered in a secure and safe manner. DevSecOps teams and practices can learn from historical governance, risk, and compliance measures and apply similar policies to AI-based code, so businesses release software created or optimized by AI in a responsible way using governance, risk, and compliance policies designed for AI-assisted software releases. And, with analytics concepts like predictive intelligence that apply AI and ML to large data sets, managers can get better insights across consolidated software delivery lifecycle information to make better trade-off decisions earlier on in the process. ” 

Taking the “I” out of AI – Inter-organizational Collaboration to Expand AI Literacy and Implementation. Commentary by Sibito H Morley III, Chief Data & Transformation Officer, Sinch 

“Currently, data scientists and software developers write and deploy an organization’s AI models as they are most well-versed in the nuances of this technology. Yet, given the widespread business impact these tools can have across the enterprise, there is debate over who should be in charge of AI efforts and what roles are needed to ensure proper and ethical adoption and implementation of these tools. 

Key technology leadership roles are evolving to incorporate AI as a key responsibility, and we’re even starting to see companies creating “Head of AI” roles to ensure a leader focused on overseeing AI across the entire lifecycle. However, other non-technical roles, such as AI ethicists, legal counsel, and business strategists, are just as critical to expanding AI literacy throughout the company and ensuring that AI is democratized and implemented responsibly. 

AI initiatives led exclusively from the technical side pose three significant risks. First, technologists can become too focused on the elegance of the solution and forget why these investments exist. This disconnect wastes valuable resources and time and creates mistrust between leadership and AI technologists. Second, AI solutions overseen only by technical teams are often unwieldy, hard to use, and fail to meet user expectations — leading to low adoption rates, dissatisfaction, and eroding trust. The third risk is societal: AI systems are at high risk of demonstrating bias, unfairness, ignoring privacy requirements, and lacking the transparency required to be called “responsible AI.” A lack of transparency can become a barrier for fellow employees who may be concerned or have unanswered questions about what is being done with AI. 

One effective approach for mitigating these risks is to create a multi-disciplinary team that guides AI strategy, initiatives, and products. Collaboration and communication between these stakeholders is vital to ensure AI solutions are technically excellent and address both business and ethical considerations. For example, risk management (often Privacy and Infosec) should provide the guardrails for evaluating potential risks from AI based on intent, data required, and utilization. This empowers the team to innovate and solve problems without exposing the company to unnecessary risk. Similarly, senior leaders and stakeholders should be responsible for guiding development efforts towards the strategic outcomes for the company and its customers. To do so, they don’t need to understand the technical details of how AI works, but they should be clear why AI matters to the business and the bottom line.” 

Digital Transformation is Dead. Commentary by Raj Sundaresan, CEO of Altimetrik

“Digital transformation as a movement fell short of expectations due to its siloed nature and a misplaced focus on technology rather than aligning with core business objectives. Digital transformation is dead. Companies have invested significant time and resources into transformation initiatives that proved to be too lengthy and ultimately failed to yield the anticipated results.

One of the key issues with digital transformation is the lack of holistic ownership, collaboration, and alignment across the entire enterprise. This disconnect hinders growth and innovation, as departments and initiatives often operate independently, resulting in fragmented efforts that fail to deliver meaningful outcomes.

Companies have instead turned their attention and investments towards Digital Business, a fresh approach based on achieving outcome-driven, incremental growth. Every industry is seeing this shift to digital business, with Fortune 500 companies and influential businesses across finance, retail, pharmaceutical, agriculture, and more implementing a digital business methodology (DBM) to unlock opportunity and growth with greater consistency, speed, and scale. 

Data democratization via a Single Source of Truth (SSOT) and AI Engine creates key insights. This rearchitecting enables enterprises to create a new operating model that empowers innovation and growth. 

The emphasis on business-led ownership, collaboration, and alignment ensures that technology is fully aligned to the business’s core objectives, resulting in a more efficient and effective transformation process. The methodology and platform underpinning digital business create a cohesive environment that nurtures collaboration and streamlined workflows, positioning organizations to respond rapidly to changing market dynamics.

By recognizing the inefficiencies and limitations of digital transformation, companies are now embracing an approach that focuses on business objectives leveraging and integrating technology and data more effectively. The newfound emphasis on collaboration, data democratization, and strategic alignment allows enterprises to thrive in an evolving digital landscape, fostering innovation, agility, and sustainable growth.

Low Code Powers Business Agility. Commentary by Varun Goswami, Vice President of Product Management at Newgen Software

“We reside in a fast-paced technological era. Enterprises today have evolved into tech hubs. The experience quality is inversely tied to time; quicker timeframes yield superior experiences (e.g., issue resolution, deliveries, information retrieval). Yet, amplifying and sustaining these experiences is immensely intricate, contingent on norms, regulations, economic fluctuations, governance, etc.  

The pressing need is to shift from business fragility to agility, ensuring quick responses to market shifts and regulations. This, however, hinges on multifaceted elements like skills, expertise, and technology.

The answer rests with low code—be it for AI, app development, or content management. Low code expedites app creation with minimal skill demands. Think of low code platforms as LEGO; the building blocks are ready; just arrange them as guided. Similar to LEGO’s varying age-appropriate sets, application development can be catered to by business users as well as pro developers. Low code speed up app development and extends experiences across mobile, portals, and in-person visits. And like one size does not fit all, low code platforms vary for different business requirements. However, the primary objective of a low code platform for faster go-to-market remains unwavering.” 

How to spot a company invested in data science career growth. Commentary by Lyndsey Padden, VP, Data Science, 84.51°

“At first glance, navigating the data science job market—whether it be for an entry or senior role—seems simple: Find a data science role at a company with a solid culture, learning opportunities, and cool work—Done! For better AND worse, today’s hiring climate is far more complex with opportunities spanning industries and across organizations in various places on the data maturity curve.

As a data science talent strategy leader, I can share my perspective on what solid organizations have in place to foster engaged and development-focused science communities. Some key indicators include: dedicated data science teams that are embedded in and across a business; ample learning and development opportunities and career mobility programs; diverse work that spans domains, science applications, or ideally both. These factors are solid indicators that an organization embeds data and data science into decisioning and sees that talent as a differentiator over a cost of doing business. No organization is perfect, but if you see these hallmarks, it’s a good sign you’ve found a nurturing environment where your career can progress.”

Enterprises will lean towards leveraging small models rather than large ones. Commentary by  Alex Ratner, CEO of Snorkel AI

“For specific tasks, specialized models consistently outperform generalized ones. Not only can they achieve higher accuracy, especially on bespoke, complex enterprise data and workloads, but their smaller architecture also makes it easier to deploy and more cost-effective, particularly at high volumes. 

Moreover, large-scale models, particularly those emerging from the open-source realm, expedite the deployment of efficient models. Data science teams can leverage these open-source models tailored to their specific domain, fine-tune them for peak accuracy, and then condense that performance into a streamlined, deployable model.”

Sign up for the free insideBIGDATA newsletter.

Join us on Twitter: https://twitter.com/InsideBigData1

Join us on LinkedIn: https://www.linkedin.com/company/insidebigdata/

Join us on Facebook: https://www.facebook.com/insideBIGDATANOW

Speak Your Mind