As AI matures this year, enterprises will start exploring and harnessing the technology and seeing ROI, according to the experts that we spoke with. From more focused AI models for specific sectors such as retail and healthcare, to AI agents transforming employee workflows, here’s what’s predicted to come for AI in the workplace in 2025. […]
As AI matures this year, enterprises will start exploring and harnessing the technology and seeing ROI, according to the experts that we spoke with. From more focused AI models for specific sectors such as retail and healthcare, to AI agents transforming employee workflows, here’s what’s predicted to come for AI in the workplace in 2025.
AI’s impact on business and operations
Josh Pitman, MD, Priory Direct
“A significant moment in 2025 will be the start of the new financial year when many measures announced in the budget come into play. The budget was a painful one for businesses and, like many others, we have been busy calculating the impact come next April to revise our recruitment plans. We need to establish what we can afford in terms of pay rises and hiring to accommodate our ongoing expansion, something we want to do in a sustainable way whilst looking after our existing teams. It will be the same story for many SMEs across the country.
“The budget has made workers a more expensive and risky investment and I question whether this could be a win for AI and automation across many sectors and spur a greater shift away from people towards robotics next year. Where budgets will be stretched, this could be a catalyst even for smaller businesses to really invest in automation in 2025 so that they can reduce their employee overheads and ultimately boost profits.”
Joel Carusone, senior vice president of data and AI, NinjaOne
“In 2025, organisations will ensure at least one person on their executive board is familiar with complex AI topics. LLMs, generative AI advances are complicated topics.
Joel Carusone, NinjaOne
And you need at least one expert in the room with intimate knowledge of the technology to answer, ‘What happens to the business when AI goes sideways?’ Especially as AI regulations become more firmly outlined and enforced, we’ll see more organisations opt for a Board model that accounts for AI expertise for businesses’ own longevity and the sake of customers’ security.”
Ofir Bloch, vice president of strategic positioning, WalkMe
“AI use by business is changing. In recent months more specialised and vertical models have been emerging. They are more reliable and accurate for their particular use cases, using a specified set of information to make them much more focused than a broader LLM. These smaller models are starting to have an impact across a range of industries— with early examples including Shopify Sensei for retail, BloombergGPT for finance, and Med-PaLM 2 for healthcare. Models also serve key departments such as HR (Paradox Olivia), customer support (Ada Chatbot) and sales (Gong AI).
Ofir Bloch, WalkMe
“We will see growing numbers of businesses adopt these more focused models next year. But just like bigger LLMs, businesses must still pay to use smaller models, and will need to ensure they are spending their money in the right areas.”
Samarth Keshava, CTO and co-founder, Sparrow
“HR teams experimenting with AI tools for leave management need to tread cautiously. 2025 will likely be the year we see the first big AI-driven leave compliance lawsuit. Employers relying too heavily on general-purpose AI tools for leave management without proper oversight are playing a dangerous game. A single error could trigger significant legal, financial and reputational risks—a reminder that you need purpose-built technology for sensitive areas like leave.”
The evolution of AI and its integration into workplaces
Karthik Suri, chief product officer, Cornerstone OnDemand
“In the gold rush of AI, distinguishing reality from hype has become increasingly challenging, leading to moments of disillusionment alongside the excitement. This year, we’ll see a clearer divide between leaders and imitators as vendors are pressed to go beyond marketing claims to deliver real solutions for real-world issues. AI is set to re-imagine the employee experience, from adaptive learning and deeply personalised coaching to in-the-flow work guidance and conversational HR platforms. Additionally, AI will offer opportunities to unlock substantial productivity in designing and administering these systems.”
“Organisations will also face greater pressure to measure the ROI of their AI initiatives, while ethical AI use and data transparency will become vital for building employee trust and attracting talent.”
Jill Goldstein, global managing partner, HR and talent transformation, IBM Consulting
“We’re entering a new chapter in how employees get work done with the rise of AI agents. Unlike AI assistants, AI agents have the ability to generate plans based on a prompt and carry out tasks independently. They are most effective when focused on specialised tasks and working together with other agents on complex, multi-part requests. As AI agents become more common, companies will need to reevaluate their work processes and create new types of teams where humans oversee groups of autonomous AI agents.”
Rodrigo Coutinho, co-founder and AI product manager, OutSystems
“For the last few years, businesses have tried to throw GenAI at all of their problems. The result: GenAI is a remarkable solution for many use cases, but not everything. As we witness the next wave of AI hype, Agentic AI, it’s important to remember that no single application of AI will be a cure-all. AI Agents have enormous potential in customer service, data consumption and analysis, and content creation. At OutSystems, we have actually created some of our own AI agents.”
Steven Webb, UK chief technology and innovation officer, Capgemini
“Gen AI will accelerate career paths for entry-level workers, and drive deeper specialisation.Our latest data shows that the technology will make many of entry-level roles more autonomous, by automating generic repetitive tasks which in turn will accelerate career progress for junior employees.
“I believe that in the next few years, we’ll also see a focus on specialisation across mid and senior management roles, driven by AI and automation. Over half of leaders and managers we surveyed (51%) believe managerial roles will shift from generalists to specialists to reflect the growing demand for deep specialist knowledge in the human-AI workforce. For instance, HR roles may transition from HR generalists to talent analytics specialists who make much greater use of AI and data.
“This shift will require organisations to reevaluate existing roles and redesign organisational structures to fully take advantage of human-AI collaboration. Responsible adoption of AI and employee training will be critical for driving successful workforce transformation in the new age of AI.”
Lucian Boldea, CEO, Honeywell Industrial Automation
“Bridging the Skills Gap: The US manufacturing skills gap could result in 2.1 million unfilled jobs by 2030. AI and automation will be critical to addressing this workforce shortage, through enhanced decision-making, issues response, and individual productivity. We can expect to see increased adoption of AI ‘digital advisors’ that will work alongside technicians and operators, enabling real-time access to data, insights and step-by-step guidance to enhance decision making. This allows less experienced workers to perform at higher levels.”
The role of AI in productivity and innovation
Lee Edwards, VP EMEA, Amplitude
“As AI moves beyond the hype cycle in 2025, organisations will shift focus from attention-grabbing new implementations to invisible, value-driving solutions that seamlessly enhance customer experiences.
Lee Edwards, Amplitude
“The AI landscape will evolve from AI features tacked onto legacy tools, toward practical applications that improve customer experiences and drive business outcomes. ‘Invisible AI’ working behind the scenes will become the new standard for successful AI implementation, separating market leaders from those still stuck demonstrating AI for AI’s sake.”
Ofir Bloch, vice president of strategic positioning, WalkMe
“2024 saw a boom in applications adding AI copilots, helping employees use software more effectively by offering contextual, real-time advice. In 2025 businesses will begin to see the risk of copilot overload, as employees get advice from different copilots, on different applications. The end result could range from mixed messages to employees being pulled in different directions, causing confusion and strategic misalignment across businesses. To get ahead of this risk, CIOs will increasingly need to guide employees through which tools they should be using, and when – ideally making it so employees don’t notice if and when they are interacting with different copilots at all.”
The changing landscape of AI regulation and oversight
Naren Narenden, chief scientist, Aerospike
“The good news is that the frenzied period—during which corporate leaders rushed to shoehorn AI into their product line—has now stopped and executives are being more considerate and realistic about their AI strategies. However, what we can expect to see now is an urgent search for researchers and technologists to find a way to reduce the cost of LLMs both financially and environmentally. Currently, the processes required for GenAI are incredibly expensive and resource-intensive, particularly when it comes to inference – the asking of a specific question—which is invoked much more frequently than the training of models.”
Eleanor Watson, IEEE member and AI ethics engineer
“According to a study by the Institute of Electrical and Electronics Engineers (IEEE), artificial intelligence (AI) is set to be the most critical technology in 2025. Most enterprise leaders (91%) believe there will be a generative AI ‘reckoning,’ as public fascination and perception is set to shift to a greater understanding of, and expectations for, what the technology can and should do. One of the key focus areas will be its ‘self-governance’ as AI will begin to independently analyse situations and produce actionable insights, without the need for human oversight.”
Finding real ROI in AI
Nick Burling, senior vice president, product, Nasuni
“2025 will usher in a more measured approach to AI investment, as organisations will be increasingly focused on quantifiable ROI. While AI can deliver immense value, its high operational costs and resource demands mean that companies need to be more selective with their AI projects. Many enterprises will find that running data-heavy applications, especially at scale, requires not just investment but careful cost management. Edge data management will be a critical component, helping businesses to optimise data flow and control expenses associated with AI.
For organisations keen on balancing innovation with budgetary constraints, cost efficiency will drive AI adoption. Enterprises will focus on using AI strategically, ensuring that every AI initiative is justified by clear, measurable returns. In 2025, we’ll see businesses embrace AI not only for its transformative potential but for how effectively it can deliver sustained, tangible value in an environment where budgets continue to be tightly scrutinised.”
Read more here: 2025 Informed: Cybersecurity and AI
