Kobi Berebi
Israeli insurtech company YuviTal has developed BID, a behavioral AI engine that combines artificial intelligence, data, and behavioral science. The technology enables insurers and healthcare organizations to better understand users’ behavioral patterns and personality-related characteristics, anticipate meaningful changes in engagement, select the most appropriate intervention, and improve the effectiveness of health, prevention, engagement, and customer retention programs.
Kobi Berebi, Deputy CEO of YuviTal, explains how the behavioral AI engine emerged from direct collaboration with clients, how YuviTal is moving beyond conventional analytics toward behavioral prediction and real-time decision support, and why the real competitive advantage in AI lies not in access to the technology, but in the quality of its implementation.
In recent years, insurers and healthcare organizations have invested significant resources in data, analytics, and artificial intelligence. The volume of available information continues to grow, and organizations will continue to require more accurate, reliable, and relevant data sources.
However, the central challenge extends beyond collecting and presenting information. The critical question is how to interpret that information, identify meaningful behavioral signals, anticipate what is likely to happen next, and translate those insights into the right decision, timely action, and measurable business impact.
In response to this challenge, YuviTal developed BID, short for Behavioral ID. BID is a behavioral AI engine that combines artificial intelligence, data, and behavioral science to help insurers and healthcare organizations better understand users’ behavioral patterns and personality-related characteristics, identify and anticipate changes in engagement, select the most appropriate intervention and incentive, and improve the effectiveness of health, prevention, engagement, and customer retention programs.
We spoke with Berebi, Deputy CEO of YuviTal, a technology company that develops platforms and solutions designed to promote healthier behaviors through incentives, personalization, and long-term behavioral change.
The conversation explored insights from working with executive teams around the world, the transition from descriptive analytics to behavioral prediction and real-time intervention, and how YuviTal connects business needs, human behavior, data, and technology to create practical, groundbreaking solutions with measurable impact.
After years of working with insurers and healthcare organizations, what insight led to the development of BID?
Berebi: Through our work with clients around the world, we understood that enriching the information available to an organization is essential and serves as a prerequisite for better decision-making. However, the real multiplier effect occurs when the organization correctly interprets that information, connects it to the relevant context, identifies what is likely to happen next, and uses it to determine the appropriate course of action.
In meetings with executive teams at insurers and healthcare organizations, the same questions arose repeatedly: How do you change the behavior of an entire population? How do you increase adherence to health programs and care pathways? How can declining engagement be identified before a member disengages completely? How can an organization determine which intervention is most likely to work for a specific individual? And where should resources be invested to generate the greatest impact?
As our discussions with clients in Israel and internationally deepened, we understood that the principal gap lies in translating an emerging behavioral picture into a practical decision.
Today, many organizations struggle to identify behavioral change as it occurs, and even fewer can anticipate it early enough to influence the outcome. Even when a change is identified, a management question remains: What should be done now, for whom, at what time, through which channel, and in what way?
BID was developed to address that need. It is a behavioral AI engine that connects artificial intelligence, data, and behavioral science to help insurers and healthcare organizations analyze behavioral patterns, identify relevant signals, anticipate likely changes, and select the next action with greater precision.
To me, this represents the right application of AI. The objective is not to add AI as an isolated feature or generate another dashboard. It is to create an organizational force multiplier that operates on an enriched information layer, processes signals from multiple sources, and generates relevant, actionable insights for stakeholders across the organizational value chain.
What problem are you seeking to solve for your clients?
Berebi: Insurers and healthcare organizations invest substantial amounts in health, prevention, remote care, customer service, and engagement programs. In many cases, however, these investments fail to generate sustained behavioral change and therefore do not fully achieve the objectives for which they were established.
The challenge is not merely to operate more programs. It is to ensure that resources are directed toward the individuals, populations, and actions where they are most likely to create value.
We aim to help clients move from managing programs to continuously managing decisions. Rather than focusing only on how many people participated, organizations need to understand who requires an intervention, what may motivate that individual, how likely the individual is to respond, when contact should be initiated, and which action is expected to generate the greatest impact relative to the investment.
For us, the path to a solution begins with understanding the business problem. That is why our work does not begin with a feature, a model, or an AI tool. It begins with a conversation with the client.
We need to understand how the organization operates, where the customer journey is disrupted, what prevents teams from acting effectively, which decisions are currently based on incomplete information, and how management defines success.
Only after understanding this reality can we design a solution that improves the user experience, supports better decisions, optimizes resource allocation, and contributes to growth and profitability.
How does BID work in practice?
Berebi: BID continuously analyzes multiple variables and behavioral signals from different data sources. These include activity patterns, engagement levels, changes over time, responses to previous interventions, interaction patterns, and other relevant sources of information.
Based on this analysis, the engine identifies which intervention may be appropriate for a particular user at a specific point in time, evaluates the likelihood that the user will respond, and determines the most effective way to engage that individual.
Analysis and prediction, however, are only the first stages. BID is connected directly to the application, enabling it to translate an insight into an actual intervention in real time.
For example, when the system identifies a consistent decline in activity and recognizes a growing risk of disengagement, it does not merely report the trend. It can select an intervention suited to that user’s behavioral patterns and activate it directly within the application.
The system then evaluates the user’s response over time and generates recommendations designed to improve the outcome of the next intervention. This creates a continuous process of learning, prediction, decision-making, intervention, and measurement.
The objective is not to send more messages or add more features. It is to select a more relevant action and deliver it at the moment when it is most likely to influence behavior.
From a business perspective, this capability enables organizations to operate with greater focus, reduce reliance on ineffective or generic interventions, and direct resources toward the populations and actions with the greatest potential to create value.
It appears that the thinking behind the product began with the clients themselves.
Berebi: Absolutely. We believe that the right way to build a product is to begin with the problem the client is trying to solve, even when advanced technology, extensive data, and significant industry experience are already available.
Over the years, we have learned that the real problem is not always the one described in the requirements document. It emerges from conversations with executive teams, service teams, professional staff, and the people responsible for operating the process day to day.
Only by understanding how an organization makes decisions, where bottlenecks arise, which information is missing at critical moments, and what the customer experiences throughout the journey can we propose a solution with a genuine likelihood of success.
That is also how I view my role. It is about taking the client’s business need, identifying the root of the problem, translating it into a clear product and technology strategy, and bringing together product, development, data, service, operations, and behavioral science professionals to build a solution that can be implemented, evaluated, and measured.
Advanced technology alone is not enough. Leadership is required to connect the technology to a business process, ensure that different teams work toward the same outcome, and maintain a clear focus on measurable value.
Where is YuviTal heading over the coming year?
Berebi: We will continue investing in BID’s capabilities, improving personalization, strengthening its behavioral prediction capabilities, and expanding the use of AI and behavioral science to support more precise decision-making.
Several capabilities with the potential to reshape how insurers and healthcare organizations understand and influence behavior are currently at different stages of development, though we cannot disclose all of them at this time.
At the same time, we are expanding our partnerships with insurers and healthcare organizations in Israel and other markets. Our objective is not simply to implement a system. It is to connect with each client’s business goals and build a solution that reflects its population, operating processes, available data, and performance metrics.
For me, success is not measured by the number of capabilities we add. It is measured by whether a new capability solves a genuine problem, improves the quality of a decision, and produces an outcome that can be measured.
In a world where adding another AI feature has become relatively easy, focus is an important management decision. Organizations must understand where additional information can improve a decision, where predictive capabilities can enable earlier action, and where the better approach is to make more effective use of the information already available.
The goal is to apply technology to improve service, optimize budget allocation, support better risk management, and contribute to growth and profitability.
Where do you believe the field is heading over the next five years?
Berebi: I believe we will see a clear transition from systems whose primary function is to present information to systems that actively analyze, anticipate, and support decision-making in real time.
Insurers and healthcare organizations will continue investing in expanding their information sources and improving data quality. At the same time, they will increasingly seek solutions that connect that information with the business context and the customer’s behavior.
They will want to understand not only what is happening, but what is likely to happen next, why it may happen, and how they should respond.
The technology solutions that lead the market will be those that combine artificial intelligence, data, behavioral science, product thinking, and a deep understanding of business. They will not be evaluated solely on the sophistication of their technology, but on their ability to help clients improve service, increase engagement, manage risk, allocate resources more effectively, and influence profitability.
In a world where AI capabilities will be broadly accessible, competitive advantage will be determined by the quality of implementation.
The organizations that lead will be those that understand their customers more deeply, identify behavioral signals earlier, recognize the specific needs of each business, use information responsibly and effectively, and translate that understanding into precise decisions and measurable action.
