Don't Just Expect More AI In 2025—Expect Better AI, Fueled By Real-Time Data

Guillaume Aymé is CEO of Lenses.io, a pioneer in the streaming market that has transformed the way engineers work with real-time data.

Agentic AI architecture was a breakout technology in 2024. This technology has progressed faster than even the more optimistic people (like me) thought possible. In prototypes, it has shown exceptional levels of potential, yet we may still be underestimating it. As always, that doesn't mean it's a shoo-in or ready at scale.

In 2025, I expect that will change, and AI agents will move from proof of concept to opening up one of the biggest disruptions our industry has ever seen.

Businesses will face exceptional pressures and demands to feed AI with data. A challenge—and opportunity—for businesses will be how real-time data pipelines are managed to get the most out of the intelligence available to engineering teams.

Businesses can expect the following data streaming trends in 2025.

The streaming-first approach will grow with AI.

As seamless automation and real-time responsiveness continue to be a priority for organizations, pivoting to a "streaming data-first" approach will drastically shift how applications and AI architectures are designed, deployed and interconnected across business domains.

AI has also become a priority for consumer-facing self-service agents, as Forrester predicts that 50% of businesses will enable the self-service help desk as their first contact touchpoint for customers in 2025.

Real-Time AI And Automation

Businesses will adopt AI solutions capable of processing and acting on real-time data to improve interactions with customers.

Replacing Traditional Data Communications

Traditional request/response communication models will give way to more asynchronous streaming-based architectures. This will simplify how the different applications in a business can communicate with each other, essential for automation.

Cross-Domain And Company Data Sharing

Businesses will rely on real-time data streaming to enable data sharing across business domains. This approach will solve disconnection issues that occur across departments, instead emphasizing collaboration, innovation and more unified business operations.

As real-time data becomes essential to deliver on an AI strategy, organizations will increasingly treat it as a product. The value of this data will explode. The ability to share or "sell" data, both internally and externally, will be a premium offering that will become more common as enterprises aim to find new ways to monetize their data.

Geopolitics and the regulatory environment will play a huge role in cloud adoption and AI strategy.

Due to constantly evolving regulations surrounding AI, AI agent providers and AI-driven energy constraints across global markets, enterprises will need to ensure the strategies they implement remain flexible, resilient and adaptive.

Hybrid Cloud Strategies For Agility

Businesses must be able to adapt to shifting market dynamics in addition to constant regulatory changes. Cloud-agnostic solutions that focus on flexibility to quickly move data and applications will enable rapid responses to these concerns.

The Rise Of Edge Computing

Edge computing will play a major role in empowering full automation with AI in remote parts of a business such as distribution centers and manufacturing plants. Relying on this approach will improve data processing across an entire centralized business network and decrease the risk of service errors.

Open-Source Technologies

Businesses will leverage more vendor-agnostic and cloud-agnostic technologies, APIs and tools, with open source being a preferred choice.

Hyper-personalization accelerates hyper-connected business.

Gartner Inc. states that "by 2027, 75% of new analytics content will be contextualized for intelligent application through GenAI, enabling a composable connection between insights and actions."

Retail merchants have already begun implementing personalized recommendations through in-store image recognition, and automotive manufacturers have pivoted toward real-time data to determine vehicle maintenance issues. It's clear that consumers continue to demand more personalized experiences, and businesses must rely on new innovative strategies like leveraging real-time data to efficiently resolve the needs and concerns of the everyday consumer.

Real-Time Data Streams

To meet personalization demands, businesses will need to accelerate hyper-connecting their IT applications and architectures across all parts of their business through real-time data streams. Relying on real-time data will help unlock previously isolated AI, analytics and software architectures, giving business leaders the ability to collaborate between systems and enabling quicker response times when addressing customer needs.

Fueling Automation

By enabling continuous automated data flow that remains connected throughout business operations, IT leaders will also fuel new innovative solutions that further meet growing personalization demands and create new levels of automation within a business.

In 2025, IT leaders should turn to real-time data streaming to unlock the potential of emerging technologies like AI and cloud solutions to meet growing consumer needs. This shift will present challenges, but it will also present opportunities. Leaders should remember that when it comes to establishing competitive advantages, it's not necessarily about getting there first but, rather, building strong foundations for long-term success.

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?