Advertisment

Interaction - Anurag Sanghai, Principal Solution Architect, Intellicus Technologies

Anurag Sanghai, Principal Solution Architect, Intellicus Technologies on the challenges and opportunities of AI for businesses in the newly emerging environment

author-image
Archana Verma
New Update
Anurag Sanghai

Anurag Sanghai, Principal Solution Architect, Intellicus Technologies

Anurag Sanghai Principal Solution Architect Intellicus Technologies on the challenges and opportunities of GenAI

Advertisment

What are the primary challenges or limitations that your company has encountered in implementing Generative AI within Business Intelligence, and how have you addressed these challenges?  

GenAI models rely heavily on the quality of the data they're trained on. Issues like inconsistent data, missing values or biases within the data can result in inaccurate or misleading insights. GenAI models can also be difficult to interpret, leading to concerns about transparency and trust in how they reached certain conclusions. Moreover, ethical considerations, especially with text-based applications, where misuse can lead to generating harmful or misleading content, also arise. Costs associated with infrastructure, like robust hardware, cloud services and high-quality data for model training, and uncertainty surrounding return on investment highlight the complexities of adopting GenAI technologies. Lastly, implementing and maintaining GenAI solutions requires specialised technical skills that may not be readily available within all organisations.  

To overcome these challenges, organisations can explore several potential solutions. Focusing on high-quality data is paramount, necessitating robust data governance practices to ensure accuracy, completeness and consistency. Employing techniques such as de-biasing algorithms can help mitigate biases in the data.  

Advertisment

Investing in explainable AI, or XAI, tools can enable the visualisation of the decision-making process of generative models and foster a better understanding of their reasoning, enhancing trust in their insights. Building internal expertise or partnering with external AI consultancies can bridge the technical skill gaps. Establishing clear frameworks for developing and deploying GenAI solutions is critical for ensuring ethical and responsible use. Implementing data encryption, access controls and clear guardrails for GenAI usage are crucial for addressing data security and privacy concerns. Custom private language models can also be considered to enhance data privacy. 

Evaluating cost-benefit trade-offs, exploring solutions like transfer learning or model compression, conducting thorough cost-benefit analyses, assessing long-term ROI and aligning GenAI projects with business goals can further contribute to cost efficiency.  

In what ways do you see Generative AI evolving in the context of Business Intelligence and Data Analytics in the near future, and how is your company preparing to stay at the forefront of these advancements? 

Advertisment

Our commitment to research and development ensures seamless integration of GenAI capabilities within the Intellicus platform. Partnerships with leading AI research institutions and niche players enable us to explore collaborative development and leverage an integrated product and solution. Community engagement through discussions and workshops ensures our solutions evolve to meet the dynamic needs of our users. Additionally, our focus on talent development includes acquiring skilled professionals and fostering internal programs to build expertise in AI, machine learning and data science, allowing us to remain at the forefront of these fields. Through these efforts, Intellicus is poised to empower users with accessible and transformative data insights, pushing the boundaries of Business Intelligence.  

How does Generative AI enhance the user experience and accessibility of your Business Intelligence and Data Analytics tools for non-technical users or those without specialised data science expertise?  

GenAI substantially improves how users interact with and access our BI tools, elevating both usability and accessibility, particularly for non-technical users or those without specialized data science expertise. Our automated insights empower non-technical users to comprehend complex data through narrative summaries and visualizations, facilitating effective communication of key findings. The integration of natural language query capabilities democratises access to data analysis by enabling users to ask questions in plain language and receive data visualisations and insights.  

Advertisment

Furthermore, automatic visualisation functionalities enable the generation of best-fit visualizations based on the data and user intent, saving time and allowing users to focus on interpreting the insights. GenAI also excels in anomaly detection by learning normal data patterns and identifying deviations, expediting root cause analysis. Our AI-backed tailored dashboards and reports deliver personalized analytics to individual user preferences and roles, enhancing user experience.  

Moreover, GenAI automates data pre-processing by identifying and handling missing values, outliers and data inconsistencies, eliminating the need for users to possess ETL expertise. Additionally, predictive analytics provided by generative models empower non-technical users to leverage predictions about future trends or customer behavior without requiring an understanding of the underlying algorithms, further enhancing accessibility and usability.  

From a broader perspective, what do you believe are the ethical considerations and potential implications of incorporating Generative AI into data analytics, and how does your company ensure ethical usage and transparency in its implementation?  

Advertisment

Incorporating GenAI into data analytics raises significant ethical considerations. GenAI models can inherit and amplify biases from the data they're trained on, leading to discriminatory or unfair outcomes. Additionally, as GenAI models are often viewed as "black boxes”, it becomes challenging to understand how they arrive at their outputs, which can erode trust and trust and make it difficult to identify and address biases. Furthermore, the ability of GenAI to generate realistic data raises concerns about misinformation and the creation of "deepfakes" that could be used to manipulate public opinion or financial markets. The potential for compromising customer information and data leakage from public language models gives rise to data security and privacy concerns. 

Intellicus leverages GenAI to transform business intelligence by generating synthetic data without risking real-world data to facilitate risk-free exploration of "what-if" scenarios and hypothesis testing, leading to deeper insights at a faster pace. We focus on XAI techniques to improve transparency, enabling users to understand the reasoning behind AI-generated insights. Moreover, we're developing features and best practices to help users assess and mitigate bias in their data before using it to train generative models. We adopt a human-in-the-loop approach, believing that AI should augment rather than replace human expertise, empowering users to critically evaluate and interpret AI-generated insights.  

Additionally, we emphasize collaboration and education by engaging in open dialogue with the data analytics community on ethical AI practices and providing educational resources to help users understand the potential benefits and risks. Through these efforts, we aim to ensure that GenAI is used ethically, responsibly and fosters trust in data-driven decisions, contributing to a future where it is a powerful tool for good in data analytics.

Advertisment

Read more from Dr Archana Verma here 

Read products news here

genai
Advertisment