AI's impact on industries in 2025

It’s hard to imagine that we started the year off at a point where generative AI was still more fantasy than fact. We’ve gone from experimentation to seeing hundreds of gen AI agents come to life in the real world. While it’s evident the gen AI revolution is in full swing — it’s still a work in progress. According to a new National Research Group study, “The ROI of Gen AI,” commissioned by Google Cloud, roughly a third of all organizations are still evaluating or testing gen AI use cases.

Gen AI is more than integrating a new technology tool, it’s a powerful business differentiator. As such, there are many considerations to take into account, from what models to choose and how to measure success to navigating cultural changes related to ways of working and organizational structure. Now, more than ever, executives are being challenged to build winning AI strategies that enable their organizations to capitalize on the current and future potential of gen AI.

A big part of this process will be recognizing the value that gen AI brings to your specific industry and understanding how it connects back to the core value chain of your business. Throughout the year, my conversations have frequently involved questions around how to identify practical use cases that add genuine business value. While executives and business leaders understand the broader applications of gen AI, which I discussed earlier this year, they want more clarity about the primary use cases being pursued in their industry and the top areas where gen AI is already boosting bottom lines.

To get answers, I asked our Google Cloud industry leaders about what’s happening now and next with gen AI across retail, healthcare, financial services, and media & entertainment. In these conversations, there were five trends that emerged: our customers are taking advantage of multimodal AI, building AI agents, using AI-powered search, building AI-powered customer experiences, and they’re working on strategies for deepfake defense.

Hopefully, sharing some of these perspectives can help shed light on the use cases bringing immediate value in your industry and the outlook for longer-term transformation in the coming months and years.

Retail

Though abundant, early gen AI retail experiments at the start of the year were often limited in value at scale and many never moved beyond the pilot stage. In the second half of the year, retailers are reshaping these initial approaches, focusing on the concrete ways gen AI can drive value for both consumers and associates. The result is a growing number of retail use cases that center on driving growth, efficiency, and optimization within the operational parts of retail business.

Mark Steel, Google Cloud’s director of retail and consumer strategy, characterizes this shift as a move from use case to business use case, saying, “Instead of only thinking about AI strategy, retailers have shifted focus to answering the question of how AI can turbocharge their business strategies. This alignment with strategic objectives is helping to apply AI to the most important priorities within organizations more broadly instead of simply testing across every possible use case and area.”

More specifically, in the near term, retailers are zeroing in on areas like customer service, marketing, and digital commerce, using gen AI search and agents to enhance existing human capability and skills. Customer service centers are deploying tools that are AI first; can automate call transcription, generate smart replies, and respond to common customer questions. In marketing functions, teams are integrating gen AI to help write briefs, brainstorm campaign concepts, and produce personalized brand content at greater scale. We also see these capabilities coming together in AI-powered customer experiences, powering personal shopping advisors, generating new product content, and creating engaging, human-like conversational interfaces to improve online shopping experiences.

All together, Steel believes these current initiatives are laying the groundwork for longer-term transformation use cases that tap into powerful multimodal models like Gemini with long context capabilities to build increasingly sophisticated AI agents that can support the efficient running and operation of their business. One example of these agents includes personal AI stylists that can combine deep customer background knowledge with a retailer’s products, promotions, and pricing to offer personalized style recommendations across multiple visits — regardless of channel or touchpoint.

“The retailers who have made the most progress to date are not only focused on technology challenges, but also what changes are needed to embrace this new opportunity,” Steel said. “AI has huge potential to fundamentally transform how retailers work, so leaders should be open to changing processes, org structures, and historical ways of working to drive maximum benefit.”

Financial services

As a historically data-rich, insight-poor industry, the financial services industry stands to gain huge benefits in the future from using gen AI to unlock value in areas like fraud detection, risk management, and customer service. However, there are many existing challenges that must be addressed around regulatory compliance, governance, and data privacy to realize this potential. As a result, we see most use cases in the industry currently focusing on cost savings and efficiency rather than new revenue generation.

Most organizations have achieved early success by using gen AI to deliver productivity gains. This includes implementing AI agents and AI-powered search to drive faster and more accurate responses to customer inquiries, accelerate incident response by detecting critical fraud alerts faster, and empower employees to work faster, smarter, and more effectively. Rather than trying to replace a function wholesale, we’re seeing financial services organizations dig into use cases that deliver more incremental productivity improvements and enable existing resources to do more.

Zac Maufe, global head of regulated industries at Google Cloud, views this trend as a reaction to the ad hoc approaches utilized during the early rush to adopt and experiment with gen AI. Many organizations are realizing they will need to have a strong understanding of how gen AI can accelerate and integrate into day-to-day tasks to maximize its impact. For example, an investment banker doing market research needs to do more than summarize information from documents with gen AI, they need to generate summaries in a specific template to truly speed up their workflow.

“What everybody’s waking up to is that gen AI is a different way of working, so how you develop this technology has to be done in a more holistic approach,” Maufe said. “Gen AI isn’t bolted on as an extra thing. To use it, you have to tailor it a bit to how people are working.”

In the long term, Maufe believes gen AI will emerge as a key building block to develop products and services that can provide financial advice and guidance, tailored for the customer. As financial institutions have increasingly transitioned towards a self-service reality, they have struggled to achieve the same amount of sales and levels of personalization previously possible when serving customers in person. Here, gen AI shows great promise for transforming customer experiences, especially online, making it easier for customers to manage accounts and navigate more complex questions like planning for retirement or selecting a mortgage loan to purchase a property.

Though exciting, Maufe notes that these transformational capabilities will take some time to build and navigate with regulatory bodies, saying, “There’s a lot of regulatory scrutiny around data privacy and explainability, and getting data into a format that's ready for use cases continues to be a challenge. Ultimately, driving revenue is more the next frontier. It’s not going to take decades, but it won’t happen overnight either.”

Healthcare

Faced with an increasingly challenging regulatory landscape, we have seen most healthcare organizations starting their gen AI journeys by tackling use cases that boost productivity and efficiency. Gen AI is ideal for assisting with routine tasks, such as appointment scheduling or processing patient intake forms, creating many opportunities to reduce the administrative workloads on clinicians.

This trend is expected to grow. As Aashima Gupta, Google Cloud’s global director of healthcare industry solutions, points out, "We expect a surge in healthcare providers and health plans adopting AI agents to assist with key administrative tasks, such as nurse handoffs, and generating easy-to-understand explanations and communications, ultimately freeing up staff for patient care and other higher-value activities."

Among healthcare organizations, the top use cases delivering near-term value are those that accelerate clinical operations. For instance, gen AI is being used to streamline back-office tasks, including clinical documentation, member and provider communications, and claims processing. AI agents can provide 24/7 support to respond to coverage queries, eligibility questions, or claim statuses. We also see healthcare providers investing heavily in leveraging gen AI tools to generate more targeted, personalized marketing outreach, from straightforward explanations and visuals for medical jargon to coding terminology used in health insurance plans.

Gupta views these initial improvements as the fundamental groundwork for greater AI transformation. Already, we see longer-term use cases taking shape that incorporate many different types of AI — not just gen AI. Some top examples include using AI to enhance access to healthcare, reimagine screening and early detection by AI-assisted image analysis in specialties like radiology and enable new innovations like AI health concierges that can offer personalized health recommendations, multilingual support, medication reminders, and care navigation advice. There’s also an increasing number of AI models emerging that simulate biological processes to aid in discovering new drugs or repurposing existing ones.

According to Gupta, the ultimate vision of AI in healthcare is to empower individuals to take control of their own health, moving us beyond simply treating diseases to actively preventing them. AI can sift through massive datasets — patient records, clinical trials, genetic information, and more — to identify patterns and predict disease risk with unprecedented accuracy. This leads to earlier interventions, more effective treatments, and the development of new therapies targeted to an individual's specific needs.

“We'll see a rise in the adoption of multimodal AI models to analyze data such as medical records, imaging data, and genomic information to draw insightful summaries, moving closer to the vision of personalized medicine. This convergence of data analysis and AI is key to unlocking a deeper understanding of individual health, paving the way for truly personalized healthcare,” Gupta said.

Media & entertainment

For years, the media and entertainment industry has described digital transformation as a shift from traditional to becoming “media-tech” companies. Now, we’re entering the next phase of that evolution — becoming “media-AI” companies. At the start of the year, we saw what one customer called a “demo-palooza” across the media and entertainment industry — individual teams experimenting with gen AI more organically without any formal, top-down approach.

Today, Albert Lai, global strategic industries director for media & entertainment at Google Cloud, says this approach is giving way to strategies with well-defined prioritization, governance, and clear business outcomes linked to the responsible use of enterprise AI.

“Many companies have been able to navigate the noise and complexity and progress from use cases to solutions, from ideation to production, and from experimentation to return-on-investment,” Lai said. “We now see use cases that span the entire media supply chain — content production, monetization, audience experience — and within the entire enterprise.”

The top use cases in the media and entertainment industry currently focus on implementing point solutions to enhance enterprise productivity — with search and AI agents — and audience personalization. For example, we see companies incorporating gen AI to automate routine customer services tasks for subscriptions and billing, accelerate back-office processes and provide better access to enterprise-wide data, drive operational and cost efficiencies for marketing, and streamline software development workflows. Gen AI is also making it easier to deliver more personalized audience experiences through multimodal recommendations and search as well as assisting with specific content localization tasks, such as dubbing and generating captions, subtitles, and audio descriptions.

In the longer term, media and entertainment organizations will likely move towards building more powerful gen AI solutions, capable of optimizing entire workflows. We’ve already seen a lot of early exploration across the full spectrum of content production tasks, from script analysis to modernizing media archive search to content understanding for moderation and monetization. Other emerging use cases include integrating gen AI into workflows to assist with storyboarding, post-production processes, and media rendering.

“Gen AI isn’t about replacing people in media and entertainment – it’s about using it responsibly to augment creative teams and to improve the viewer experience while helping to save costs and time and to grow revenue,” Lai said.

Finding the right AI use cases for your business

Overall, the last several months have been eye-opening, revealing new lessons and takeaways about the execution of gen AI. In the year ahead, I anticipate that we’ll start to see real differentiation from organizations that are able to launch AI strategies that make the most of their investments. Much of that work will be around understanding where the biggest opportunities are in your industry and prioritizing the right use cases to make an impact. But the rise of generative AI also brings new challenges.

As AI becomes more sophisticated, so too do the methods of malicious actors. Deepfakes, AI-generated synthetic media that can convincingly replace someone's likeness or voice, pose a growing threat to individuals and organizations alike. To combat this, organizations must prioritize "deepfake defense" as a key component of their AI strategy. This involves investing in technologies and strategies to detect and mitigate the spread of deepfakes, protecting their brand reputation and fostering trust in an increasingly digital world.

At Google Cloud, we are committed to sharing our decades of experience with pioneering AI innovation and helping organizations navigate through their toughest challenges. In particular, we are developing clear guidance and frameworks on how leaders can approach and develop robust AI strategies within their organization. For instance, we use a simple prioritization matrix to help us plot out potential use cases, comparing their expected value generation against their estimated actionability and feasibility, to help customers assess their capabilities and develop their AI roadmaps.

Generative AI has quickly moved beyond being a novel concept and is now a dynamic tool for achieving business objectives and transforming industries. Looking ahead to 2025, businesses must identify their core value drivers and explore how generative AI can reimagine processes and rebuild experiences to fully realize the return on AI investments.

Download our full report, 2025 AI Trends here.