Another year, another investment in artificial intelligence (AI). That has certainly been the case for 2024, but will the same momentum continue for 2025 as many organizations begin to question its ROI?
According to most analysts, the answer is an overwhelming yes with global investment expected to surge by around a third in the coming 12 months and continue on the same trajectory until 2028. However, while budgets may be increasing, I see a more caution approach in 2025 with companies becoming discerning about the type of technology they need, and more importantly, if it can overcome specific real life business challenges.
With that said, here are some of my predictions for 2025:
1. Better Analysis Before Taking the Plunge
With more emphasis on improved ROI, businesses will be turning to AI itself to ensure they are spending wisely. One of the biggest problems to date is the haste to “jump on the bandwagon” especially since the introduction of generative AI and LLMs. In fact, as many as 63% of global business leaders admit their investment in AI was down to FOMO (fear of missing out), according to a recent study. This is why a data driven approach is essential. Following on agentic automation, cognitive process intelligence will focus on providing deeper context around business operations, essentially giving AI the capability to act as an operational consultant. These systems will be able to map, analyze, and predict complex workflows within an organization, then recommend improvements based on real-time data analysis and past patterns, beyond simple task automation. This will appeal especially to sectors like finance, logistics, and manufacturing, where even minor improvements in operations will translate into significant cost savings.
2. The AI-First Era Renews Interest in BPM
A new golden age of business process management (BPM) is on the horizon. Not since the 1990s, when the emergence of enterprise resource planning (ERP) sparked widespread digitization, have companies needed to revisit how they operate to stay competitive. Two factors are driving the change. First, companies realize that growth at all costs is not sustainable with a shift toward performance and efficiency to achieve healthy unit economics and positive ROI. Second, the gen AI agentic hype accelerated interest and adoption of the technology as company executives mandated teams to explore use cases, looking to gain market advantages.
The most effective model or intricate prompt is unproductive in isolation. As a result, BPM is once again in the limelight. AI’s imminent influence on almost all enterprise workflows makes process discovery, analysis and redesign fundamental for operationalizing any program, let alone scaling it. This predicament mirrors previous digital transformation challenges, which suffered poor success rates due to excessive technology focus while neglecting human or process considerations.
3. More Integrated Multimodal AI Systems
Multimodal AI that combines text, vision, audio, and sensor data will become the norm for businesses seeking holistic, situational awareness. This will go beyond standalone document analysis or voice recognition; instead, integrated systems will be able to draw insights from multiple modalities to provide richer, more accurate interpretations of complex scenarios.
In the financial sector, multimodal AI can revolutionize customer service by integrating text, voice, transaction records, and behavioral data to provide a comprehensive understanding of customer needs. This integration enables financial institutions to offer personalized services, enhance customer satisfaction, and improve operational efficiency.
For instance, AI-powered virtual financial advisors can provide 24/7 access to financial advice, analyzing customer spending patterns and offering personalized budgeting tips. Additionally, AI-driven chatbots can handle high volumes of routine inquiries, streamlining operations and keeping customers engaged.
By leveraging multimodal AI, financial institutions can anticipate customer needs, proactively address issues, and deliver tailored financial advice, thereby strengthening customer relationships and gaining a competitive edge in the market.
4. Regulation-Ready, Explainable AI
With global regulations on the rise, there will be a focus on explainable and transparent AI that meets regulatory requirements from the ground up. We’ll see more emphasis on tools that enable AI transparency, bias reduction, and audit trails, allowing companies to trust their AI solutions and verify compliance on demand.
AI developers will likely provide interfaces that allow stakeholders to interpret and challenge AI decisions, especially in critical sectors like finance, insurance, healthcare, and law.
Beyond transparency, a commitment to responsible AI will be a priority as companies try to gain the trust of clients and consumers. The OECD reports over 700 regulatory initiatives in development across more than 60 countries. While legislation is still catching up to innovation, companies will be seeking to proactively follow voluntary codes of conduct, like those developed by IEEE or NIST, to establish clear standards. By embracing transparency, adhering to best practices, and clearly communicating with customers, they foster a reputation for reliability that bridges the trust gap in AI and increases loyalty and confidence.
External audits will also grow in popularity to provide an impartial perspective. An example of this is forHumanity a not-for-profit organization that can provide independent auditing of AI systems to analyze risk.
5. Human-Centered AI Design
As AI tools become more embedded in our lives, ethical considerations and human-centered AI design will grow in importance. Expect to see a shift toward AI systems designed with a humanistic approach, prioritizing user empowerment, inclusivity, and well-being.
Companies will likely aim to develop AI solutions that emphasize collaborative intelligence—AI systems that enhance human decision-making rather than replace it. This might also include a focus on psychological safety and user well-being in human-machine interactions
6. Hold your Horses Agentic
The boundaries between deterministic and agentic automation will blur in 2025, leading to more integrated, intelligent, and adaptive systems that enhance various aspects of our lives and industries. But deterministic automation will continue to rule and power at least 95% of automation in production next year.
No doubt agentic automation, characterized by systems that can make autonomous decisions and adapt to new situations, is sexy and poised to make substantial strides. In dynamic environments where flexibility and adaptability are crucial, these systems will enable more personalized and responsive interactions, improving user experiences and outcomes.
7. Pushback on LLMs
The advancements in large language models (LLMs) have been nothing short of revolutionary. But, as with all great things, they come with their own set of challenges, notably the hefty price tag on resources.
Many drawbacks of generative AI and LLMs stem from the massive stores of data that must be navigated to yield value. Not only does this raise risks in the way of ethics, accuracy, such as hallucinations, and privacy, but it grossly exacerbates the amount of energy required to use the tools.
Instead of highly general AI tools, 2025 will see enterprises pivot to purpose-built AI specialized for narrower tasks and goals. It’s like chopping back what you don’t really need – just like a Bonzi tree – you have to cut it away, so it becomes leaner and more efficient. By compressing the model itself, the precisions of its calculations are smaller, increasing speed and lowering energy requirements for computer power.
Wrap up
Without a doubt, 2025 will be another year of greater investment in artificial intelligence, particularly generative AI which will continue to transform companies and jobs in every sector. However, business leaders will take a more data-driven, holistic approach to investment that achieves real business goals, while also ensuring standards are met on ethics and sustainability. After all, the real potential of AI is found in the way it is thoughtfully and strategically applied – don’t let FOMO cloud your judgement.
Credit: Source link
