Large language models (LLMs) are increasingly applied to multi-modal data analysis--not necessarily because they offer the most precise answers, but because they provide fluent, flexible interfaces for interpreting complex inputs. Y et this fluency often conceals a deeper structural failure: the prevailing "Prompt-to-Answer" paradigm treats LLMs as black-box analysts, collapsing evidence, reasoning, and conclusions into a single, opaque response. The result is brittle, unverifiable, and frequently misleading. We argue for a fundamental shift: from generation to structured extraction, from monolithic prompts to modular, agent-based workflows. LLMs should not serve as oracles, but as collaborators--specialized in tasks like extraction, translation, and linkage--embedded within transparent workflows that enable step-by-step reasoning and verification. We propose Data-Puzzle, a conceptual multi-agent framework that decomposes complex questions, structures information into interpretable forms (e.g., tables, graphs), and coordinates agent roles to support transparent and verifiable analysis. This framework serves as an aspirational blueprint for restoring visibility and control in LLMdriven analytics--transforming opaque answers into traceable processes, and brittle fluency into accountable insight. This is not a marginal refinement; it is a call to reimagine how we build trustworthy, auditable analytic systems in the era of large language models. Structure is not a constraint--it is the path to clarity.

The rise of Large Language Models (LLMs) and generative visual analytics systems has transformed data-driven insights, yet significant challenges persist in accurately interpreting users' analytical and interaction intents. While language inputs offer flexibility, they often lack precision, making the expression of complex intents inefficient, error-prone, and time-intensive. To address these limitations, we investigate the design space of multimodal interactions for generative visual analytics through a literature review and pilot brainstorming sessions. Building on these insights, we introduce a highly extensible workflow that integrates multiple LLM agents for intent inference and visualization generation. We develop InterChat, a generative visual analytics system that combines direct manipulation of visual elements with natural language inputs. This integration enables precise intent communication and supports progressive, visually driven exploratory data analyses. By employing effective prompt engineering, and contextual interaction linking, alongside intuitive visualization and interaction designs, InterChat bridges the gap between user interactions and LLM-driven visualizations, enhancing both interpretability and usability. Extensive evaluations, including two usage scenarios, a user study, and expert feedback, demonstrate the effectiveness of InterChat. Results show significant improvements in the accuracy and efficiency of handling complex visual analytics tasks, highlighting the potential of multimodal interactions to redefine user engagement and analytical depth in generative visual analytics.

User trust is a crucial consideration in designing robust visual analytics systems that can guide users to reasonably sound conclusions despite inevitable biases and other uncertainties introduced by the human, the machine, and the data sources which paint the canvas upon which knowledge emerges. A multitude of factors emerge upon studied consideration which introduce considerable complexity and exacerbate our understanding of how trust relationships evolve in visual analytics systems, much as they do in intelligent sociotechnical systems. A visual analytics system, however, does not by its nature provoke exactly the same phenomena as its simpler cousins, nor are the phenomena necessarily of the same exact kind. Regardless, both application domains present the same root causes from which the need for trustworthiness arises: Uncertainty and the assumption of risk. In addition, visual analytics systems, even more than the intelligent systems which (traditionally) tend to be closed to direct human input and direction during processing, are influenced by a multitude of cognitive biases that further exacerbate an accounting of the uncertainties that may afflict the user's confidence, and ultimately trust in the system. In this article we argue that accounting for the propagation of uncertainty from data sources all the way through extraction of information and hypothesis testing is necessary to understand how user trust in a visual analytics system evolves over its lifecycle, and that the analyst's selection of visualization parameters affords us a simple means to capture the interactions between uncertainty and cognitive bias as a function of the attributes of the search tasks the analyst executes while evaluating explanations. We sample a broad cross-section of the literature from visual analytics, human cognitive theory, and uncertainty, and attempt to synthesize a useful perspective.

Every person interacts with some device or application that is AI-enabled at least once a day. That is how widespread AI is today. Every time you perform a voice search using Siri or Google Assistant, you are leveraging the power of artificial intelligence. As AI continues to evolve and improve, it will find more and more practical use in everyday business operations. Moving into 2022, AI will play a pivotal role in influencing digital transformation across industries.

Industries across the enterprise are advancing their use of both private and public 5G networks as an increasing number of CIOs and other leaders identify business opportunities that require the capacity, low latency and reliability that only the fifth generation of broadband cellular technology delivers. Recent research underscores that growth. The global 5G services market is expected to become a $664.75 billion market, according to a recent 5G market report from Grand View Research. That means a compound annual growth rate of 46.2% from 2021 to 2028. The research firm also predicted 5G adoption and use to grow in numerous industries, including agriculture, retail and utilities. CIOs and other executives across verticals will need to identify the business opportunities 5G enables and understand where and how it can be the differentiator.

At Taiho Oncology, we work with multiple CROs across our spectrum of clinical trials – a common scenario for many pharma companies. As a result, in 2019 we also had multiple sources of clinical data, including electronic data capture (EDC), clinical trial management system (CTMS), lab data, etc., that were siloed, difficult to access, and not being leveraged to their fullest potential. This scenario was unsustainable and suboptimal for our company, customers, and patients. Aggregating that complex and voluminous data into a single source of truth at the right frequency to better inform business decision-making and collaboration was a clear priority for us. We tried multiple single-point solutions, none of which yielded the results we required.

In this developer code pattern, we will analyze insurance claims data and determine whether there are any fraudulent claims filed by users. We do this by analyzing data patterns using IBM Db2 Graph. Analysts from insurance companies can visually analyze the graph by finding patterns in data related to patients, doctor visits, multiple claims, etc. and determine whether there are suspicious claims filed. As the volume of data grows, it has become a challenge to analyze vast networks of connected data. To overcome this, there is a rapid adoption to graph database technologies, since they're built around relationships and represent data in a way that is more intuitive to read and gain insights.

If you think AI and analytics are better apart, you may be missing out on some valuable opportunities. I was shocked recently while gathering background information for a short e-book I co-authored with Ellen Friedman, AI and Analytics at Scale: Lessons from Real-World Production Systems, to find that some people still think large-scale analytics projects and AI projects should be siloed and segregated. In particular, these people think of AI systems as very expensive, very specialized, separate systems, that must be completely isolated from analytics systems. My take is just the opposite. For years we've observed real-world enterprises across many sectors that benefit by co-locating analytics – even legacy analytics – together with modern AI and machine learning projects.

No Thanks, We Have Data. Everywhere we are told that data makes winners and winners use data. No "animal spirits" for us, thank you very much! But abandon hard-won good judgment? There was a time when decision makers were criticized for "analysis paralysis", which is the failure to choose.

Over the past several years, technology has rapidly changed what enterprise analytics can do. Analytical approaches that incorporate predictive models have begun to displace merely descriptive approaches. Descriptive analytics, which continue to be valuable for many users, have evolved as well, making greater use of visual analytics and moving toward a self-service model in which nontechnical users can often develop their own analyses. In general, analytics are quickly becoming both easier to use and more powerful. Despite this progress, it's still difficult to use data and analytics to understand and predict many of the important phenomena in organizations. Predictive models require a substantial amount of past data and a reasonable amount of expertise to create and use, which limits how and when they can be deployed.