Once seen as a futuristic concept, AI has become a powerful economic force, reshaping industries and redefining global competitiveness.

According to an IDC study, business investment in AI is set to generate nearly $20 trillion in economic impact by 2030. This rapid adoption—driven by the pursuit of operational efficiency and enhanced products and services—is expected to boost global GDP by 3.5% by the end of the decade.

The momentum is undeniable. 98% of business leaders now rank AI as a top priority, highlighting its transformative potential across sectors.

Even in the Middle East, the impact is set to be profound. McKinsey estimates that generative AI alone could contribute between $21 billion and $35 billion annually to GCC economies—on top of the already substantial $150 billion potential from other AI technologies. This represents a significant share of the region’s non-oil GDP, signaling a major economic shift.

The race is on. Early adoption and AI readiness are now critical for nations and businesses looking to seize the opportunities of this transformative era.

The question is no longer whether AI will reshape the global economy but who will lead the way and who will struggle to keep up.

REAL INNOVATION AMID TRENDS

Shawn Edwards, Chief Technology Officer at Bloomberg LP, believes one of the biggest challenges of leading innovation is maintaining a pragmatic approach.

“This means I need to assess which emerging technologies are valuable for Bloomberg to adopt and which aren’t worth their hype. I often joke that there are more fads in the tech sector than in the fashion industry,” Edwards notes.

He stresses the need to harness new technologies to solve real business challenges while focusing on meaningful impact. Equally crucial, he notes, is adaptability—quickly exploring and implementing emerging innovations to stay ahead in an ever-evolving industry.

“Generative AI illustrates all of these challenges. It is not a silly fad. It’s one of the most exciting advancements we’ve seen in decades and we’re already finding this incredibly powerful tool useful in real products,” Edwards says. “However, it is more complicated to properly use and scale in products.”

Edwards explains that Bloomberg’s mission is to bring transparency to financial markets, and AI plays a crucial role in managing vast amounts of data and the accelerating flow of information. As the volume of financial news surges, so do the challenges for professionals navigating this influx.

For equity analysts and portfolio managers, the need to extract insights quickly—especially from lengthy earnings transcripts—has never been greater. In a volatile market with rising news volumes, Bloomberg leverages generative AI to distill stories into three key points, enabling users to pinpoint relevant information faster and make informed decisions more efficiently.

“These are the kinds of real problems that AI can be – and should be – helpful with.”

THE ROLE OF AI IN FINANCE

Edwards highlights that discussions around AI and machine learning in financial data often focus on their role in developing AI models. However, an equally important aspect is how AI is used to extract and process data—something Bloomberg has been implementing for years.

“Our data analysts used to manually copy numbers from earnings tables by hand into databases. Now, Bloomberg’s AI systems are essential to the fast and accurate collection of the company-reported data and broker estimates that power our best-in-class company analysis functionality.”

Bloomberg employs natural language processing, machine learning, and computer vision to extract and enhance key financial data from text, tables, and documents. These technologies also generate actionable insights from images, audio, blogs, and social media, expanding the depth and breadth of market intelligence.

These AI-driven workflows power Bloomberg Terminal functions and ensure fast, accurate data delivery. Achieving this level of precision demands rigorous model training, fine-tuning, and continuous optimization to enhance speed and reliability.

“We are constantly looking at ways to further expand the application of our extraction pipelines while improving their performance as new AI techniques emerge.”

RELIABILITY IS KEY

Finance is an industry that relies on facts to make mission-critical decisions,” Edward states “So, one crucial principle we think about when designing solutions using Generative AI is that we derive answers from trusted sources. This includes, for example, the 40 years of time-series data Bloomberg has cleaned, the hundreds of millions of documents (e.g., press releases, earnings call transcripts, bond prospectuses) that we’ve collected, and the news stories our journalists have written.”

Bloomberg prioritizes transparent attribution, allowing users to access the original data or sources behind AI-generated outputs. Clear disclaimers indicate where AI is applied. This approach keeps users engaged with the original source, ensuring they have enough context to verify accuracy. 

Additionally, financial professionals can consult document authors for further insights. Built-in verification steps help prevent errors, reinforcing a cautious approach that ensures reliability.

“Context is everything, and we also make it easy for users to flag incorrect or inaccurate answers so that we can address them promptly,” Edwards states.

“Guiding principles like this one help us ensure that outputs are trustworthy and reflect reality. This isn’t just the right thing to do. It’s what our clients demand. They rely on us for accurate insights, and we must uphold this trust,” he adds.

FUTURE OPPORTUNITIES

Edwards is incredibly bullish about the capabilities of Generative AI, which can help people interact more naturally with data.

“In particular, we see an opportunity to make Bloomberg Query Language (BQL) even easier to use and Bloomberg data even more accessible. BQL is already a powerful data and analytics engine that enables our customers to retrieve and analyze normalized, curated, point-in-time Bloomberg data across multiple asset classes, each with its own fields, functions, and parameters,” he says. 

“Generative AI will enable us to make BQL more accessible by transforming natural language queries into valid BQL queries.”