Franklin Templeton chief urges businesses to adopt AI faster and protect core data

Franklin Templeton's Jenny Johnson called on companies not to delay the implementation of new technologies and to more actively adopt artificial intelligence. In her logic, the question is no longer about the fashion for AI, but about how quickly business can embed it into real processes without losing control over key data.

Why AI Has Become Mandatory

Speaking on Bloomberg Surveillance in Washington, Johnson essentially articulated a position that increasingly major corporations and investors share: AI can no longer be held in the status of a laboratory experiment. For executives, it is already not a separate innovation program, but a working tool that affects productivity, decision-making speed, and operational costs. This is especially noticeable in sectors with a lot of manual analytics, reporting, and customer service, including finance, consulting, and corporate services.

"Companies need to learn to use new technologies, including AI."

The essence of her thesis is not that any business must urgently buy fashionable tools. The point is different: companies that learn to properly use new systems ahead of competitors will gain a cumulative effect. They process data faster, better understand customer behavior, and scale internal processes more cheaply. Those who continue to treat AI as optional for a separate team risk finding within a year or two that their lag has become not marketing, but operational and financial.

Data as the Foundation of Implementation

Johnson's second important emphasis is the protection of core data—the basic data on which the company's operations depend. This is especially important against the backdrop of the rapid deployment of generative models, which need access to internal documents, customer information, analytics, and correspondence. If a business gives AI access to poorly labeled or inadequately protected data arrays, it creates a double risk for itself: leaks and incorrect answers. In such a configuration, even a powerful model will not provide reliable results, because the problem will not be in the algorithm, but in the foundation of the entire system.

In practice, this means that AI implementation does not begin with a beautiful demonstration for the board of directors, but with infrastructure work that rarely looks impressive but determines the final result. Companies need to understand in advance which data can be connected to models, which data cannot be sent to external services, who is responsible for source quality, and how results are verified. Without this, even a successful pilot will quickly run into questions of security, compliance, and trust from employees.

• inventory and classification of key data
• differentiation of access for employees and external services
• connecting models only to verified sources
• measuring impact through time, quality, and risk reduction

For a financial company, this is an almost obvious rule, but now it is becoming universal for other industries. If a company does not understand which data is truly critical, who uses it, and where it is stored, the conversation about AI remains superficial. Mass deployment of models without data discipline may temporarily accelerate individual tasks, but then bring costly errors in reporting, support, compliance, and decision-making.

The Market Looks Wider

Johnson also touched on consumer conditions and oil prices—on the surface these are separate topics, but for a management company of such scale they are directly related to the conversation about technology. Consumer demand affects business revenue and companies' appetite for investment, while oil remains one of the markers of inflationary pressure and costs. Therefore, the question of AI for major executives has long not lived in a vacuum: it is considered together with expenses, rates, demand, and business model sustainability. Hence the more sober approach to implementing new systems in 2026.

The board of directors wants to understand not only where processes can be automated, but also how this solution will behave when the macroeconomic environment changes. If the consumer weakens, companies need faster control over margin and demand. If energy prices rise, the cost of inefficiency increases. In such an environment, AI becomes not a window display of innovation, but a way to make decisions faster in conditions where an error is more costly than before.

What This Means

The signal from Franklin Templeton is simple: business can no longer just discuss AI at strategic sessions, it needs to bring it to processes, metrics, and protected work with data. The winners will not be those who talk loudest about transformation, but those who link technologies with operational discipline and know how to make decisions against the backdrop of changing demand, costs, and prices.