![Industry Perspectives Industry Perspectives](https://eu-images.contentstack.com/v3/assets/blt07f68461ccd75245/bltcf2b437471bc0d1c/664ba3979c1fd769482032c6/IndustryPerspectives_ITP_Light.png?width=700&auto=webp&quality=80&disable=upscale)
Insight and analysis on the information technology space from industry thought leaders.
2024's Strategic Imperatives for Digital Transformation
With the emergence of new technologies including generative AI, here are four best practices for navigating the digital horizon.
January 25, 2024
![digital transformation key digital transformation key](https://eu-images.contentstack.com/v3/assets/blt07f68461ccd75245/blt629c263103f46025/6617f8c43340fa7946e346fc/digital-transformation-key_0.jpg?width=1280&auto=webp&quality=95&format=jpg&disable=upscale)
Amidst worldwide macroeconomic uncertainties, emerging technologies are critically changing the dynamics of achieving business goals. C-suite executive leaders are boldly building forward-thinking business use cases to create a roadmap for their AI vision. Based on large language models (LLMs), modern content generation AI tools include ChatGPT, Bard, Copilot, and LLaMA, as well as creative text-to-image AI systems, such as DALL-E, Stable Diffusion, and Midjourney. ChatGPT turned 1 years old in November 2023, even as organizations strive to adopt it for business value creation.
By 2032, the generative AI market is expected to grow to USD 1.3 trillion. Tech hyperscalers, like Google Cloud, Microsoft, Oracle, and Amazon Web Services, are now offering next-gen AI platforms for enterprise clients to develop, train, and operate their own models to realize the full potential of generative AI. With the emergence of new technologies and their worldwide impact, enterprises must identify their priorities for strategic digital transformation in 2024.
Prioritizing Generative AI to Recalibrate Business Models
Technology and service providers are upgrading their technology stacks to match the full potential of AI and generative AI for internal and external value creation.
C-suite executives now consider GenAI to be of utmost prominence as it drives innovation across a range of industries, including life sciences, healthcare, legal, financial services, and the public sector. By 2026, more than 80% of organizations will be using generative AI-powered application programming interfaces (APIs) and models or have deployed generative AI-enabled applications in production workflows.
AI is a powerful ally for faster software building tools, aiding the software development community beyond mere code generation. AI-assisted ML tools, such as AutoML (automated machine learning), TensorFlow, Amazon SageMaker, and no-code app development platforms, attribute advanced learning algorithms like deep learning to train ML models to facilitate faster ML software development with minimal coding skills.
Technology also plays a pivotal role in identifying any bugs or errors in a code and predicting future outcomes. It is a powerful tool for software testing and quality assurance. Generative AI eliminates rote writing in code generation, such as writing boilerplate codes or repetitive codes. A study conducted by McKinsey reports that software developers completed their coding tasks twice as fast using generative AI. Today, there are dedicated tools, such as GitHub Copilot, Tabnine, CodeWP, and Amazon CodeWhisperer, which can seamlessly fit into the preferred integrated development environment of a coder.
Generative AI, however, brings risks along with benefits. Since generative AI content creation tools are based on large language models, they may fall prey to bias or "hallucinations" and sometimes present fabricated information as truth.
Charting the Rise of Vertical SaaS
Vertical SaaS, an iteration of software-as-a-service, is on the rise, with market projections reaching over USD 157 billion by 2025. In 2024, enterprises are expected to adopt vertical SaaS for industry-specific niche SaaS solutions. In contrast to horizontal SaaS, which solves generic problems and targets a wider audience, vertical SaaS involves implementing industry-specific, cost-effective, and tailored software solutions to address clients' unique challenges.
An ideal use case of vertical SaaS in the healthcare industry includes features, such as electronic health records (EHRs), telemedicine capabilities, and a Medicare enrollment dashboard, where it is critical to manage patient data, medical records, and regulatory compliance details. Supply chain optimization and project management tools might be a solution that can be considered for the construction industry.
Globally, regulatory bodies in healthcare are collaboratively working to establish standards for healthcare interoperability with control over their health information. The effort involves encouraging the adoption of standards-based APIs.
Vertical SaaS providers play a pivotal role in addressing the growing significance of interoperability in healthcare. They can collaborate with healthcare providers to establish connections among various systems, including EHRs, clearinghouses, billing vendors, remote patient monitoring tools, and more. This connectivity streamlines financial workflows, facilitates care management, and improves outcomes for remote patients. Players can also leverage these tools to access EHR and patient intake data across providers, leading to enhanced prior authorization rates, identification of care gaps, and the retrieval of medical records. Furthermore, these APIs empower software companies serving healthcare organizations by providing access to clinical and financial data across diverse platforms.
As interoperability standards progress and the adoption of value-based care gains momentum, these APIs become indispensable tools for healthcare systems to align with national standards and boost profitability through value-based reimbursements. Adopting a strategic approach to software development in vertical SaaS can result in enhanced customer satisfaction, a faster go-to-market, and business success over the long term.
Using Small Language Models
Large text datasets are used to train language models and artificial intelligence systems that can perform tasks, including text generation, document summarization, language translation, and question answering. Essentially, the same function is performed by small language models (SLMs) but with noticeably smaller model sizes.
In general, researchers see language models with less than 100 million parameters as relatively small, and some even stop at much smaller thresholds, such as 10 million or 1 million parameters. SLM is a compact version of LLM, however, built for context-specific business requirements. SLM is a suitable playground for developers and application architects to develop customized, business-specific, GenAI-enabled tools using enterprise-oriented datasets.
One example is from Microsoft's Phi-2, a 2.7 billion-parameter transformer-based language model with advanced prediction abilities that is trained on a mixture of synthetic and web datasets for natural language processing (NLP) and coding.
Preventing Data Leaks
With the transformation in business use cases, governance, data privacy, and regulatory compliance are addressing cyber threats. According to a McKinsey study, nearly 53% of enterprises recognize cybersecurity as a concern associated with GenAI, but only 38% are taking steps to reduce that risk. As the foundation models of generative AI are being trained and applied to the volume of use cases every single minute, data breaches are surfacing periodically. Accessing unauthorized data from unreliable public sources is the root cause of privacy issues and data leaks in generative AI.
Organizations are fine-tuning their risk modeling methods, lowering risk scores, and identifying the factors of risk exposure. Also, we may see global legal enforcement on data privacy and compliance regulations for generative AI. For digital and cloud transformation service providers, it is crucial they enhance their cybersecurity offerings with flexible deployment capabilities.
As organizations focus on strategic digital transformation in 2024, they must prioritize generative AI to recalibrate business models, while adopting a disciplined approach to governance, security, and regulatory compliance. A proactive approach to mitigating cybersecurity risks, refining risk modeling methods, and complying with evolving data privacy regulations will be essential for the sustained success of enterprises navigating the transformative journey fueled by generative AI.
Raman Sapra is President and Global Chief Growth Officer at Mastek.
About the Author
You May Also Like