2025 was a pivotal year in which AI was even more influential on the technology sector than in previous years.
Over the past 12 months, advanced AI systems have been integrated into digital infrastructure, business workflows, cybersecurity and consumer products, with trends such as generative AI, agentic AI, hyperautomation and AI-powered cybersecurity beginning to redefine the tech landscape.
AI governance and new regulatory frameworks have continued to evolve, with the gradual implementation of the EU’s AI Act and new international initiatives defining approaches to transparency, safety and accountability.
As the new year approaches, the editorial team at National Technology News interviewed industry and policy experts to hear their predictions on the technology innovations and trends that will dominate 2026.
AI moves from hype to accountability
Experts agree that 2026 will mark the end of large-scale AI experimentation without results. After years of pilot projects, organisations will be under pressure to deliver measurable ROI, shifting the focus from flashy demos to AI that improves KPIs and delivers tangible business results.
Kai Wombacher, product manager at IBM Kubecost says: “If 2025 was the year of exponential AI investment, 2026 will be the year of marrying cost optimisation into development practices.”
As organisations face increasing pressure to demonstrate the value of AI and align investments with business outcomes, FinOps for AI will also gain importance as a fundamental framework for understanding and optimising AI and infrastructure spending, adds Wombacher.
Eugene Khostov, chief product officer at Apptio, agrees, predicting that next year will bring business back into context and value back into the IT conversation.
“The cost and complexity of the enterprise IT estate is a fact,” explains Khostov. “But the opportunity to significantly increase visibility, efficiency and effectiveness and translate into business value for stakeholders remains a white space for organisations.”
Because of this, he predicts that more enterprises will shift from project-based to product-based operating models where funding and resources are no longer tied to static, annual plans, but instead track product performance and augment in real time.
AI-driven cyber threats become the dominant security risk
AI-powered attacks, including deepfakes, AI-driven social engineering, and AI-generated malware, will emerge as a major concern for organisations. This will lead business leaders to strive to be better prepared to deal with AI-driven threats, Chris Dimitriadis, chief global strategy officer at IT governance association ISACA.
Research from the organisation finds that 51 per cent of European IT and cybersecurity professionals believe that AI-based cyber threats and deepfakes will keep them awake at night in 2026.
“Only 14 per cent of respondents feel their organisation is very prepared to manage the risks associated with generative AI solutions in 2026,” Dimitriadis said.
He points out that in the year ahead, businesses will have to adopt AI governance frameworks and invest in training focused on the responsible and secure use of AI.
Recent high-profile cyber-attacks, such as those against Marks & Spencer and Jaguar Land Rover, have also demonstrated the critical importance of business continuity, pushing many companies to focus on upgrading their technology stack to address potential new challenges.
“In the year ahead, organisations will focus on adopting resilience standards and cyber certifications, implementing stronger cybersecurity frameworks and building robust incident response plans,” Dimitriadis says.
Trust, governance and regulation will become competitive advantages
With regulations such as the EU AI Act, NIS2, DORA and the UK Cyber Bill coming into force, governance will no longer be a mere bureaucratic exercise. Experts agree that by 2026, leading organisations will integrate ethics, transparency, explainability and regulatory intelligence directly into their systems and workflows, using trust and compliance as differentiators rather than barriers to innovation.
Rebecca Crook, chief executive at digital impact company MSQ DX, says: “Mass AI adoption has defined the last few years and regulation has lagged behind, but in 2026 that gap is going to close.”
Crook adds that EU AI regulations will come into force in August, with British companies serving European customers needing to take note of the regulatory changes or face exposure to risks.
She says that the impact of the EU AI Act will be particularly significant for businesses with responsibility for digital products.
“For UK businesses, if your AI system affects EU citizens whether through personalisation engines, customer targeting or automated decision making, you are in scope,” she explains.
Crook goes on to say that organisations best positioned to navigate these changes are acting now by auditing their AI stacks, identifying high-risk systems, documenting how algorithms make decisions, and implementing meaningful human oversight.
“Equally important is getting data governance right, and remembering AI cannot legally operate without robust consent frameworks in place,” Crook adds.
Cathy Mauzaize, president Europe Middle East and Africa (EMEA) at American software company ServiceNow echoes this view: “The EU AI Act marks a turning point from theory to practice. But rules alone won’t create responsible AI, the real test is how organisations embed accountability and transparency into everyday workflows, data and decisions.”
She cites Oxford University’s 2025 Annual Report on AI Governance, which states that organisations are already integrating governance directly into workflows, rather than treating it as a mere compliance exercise.
“Leaders who succeed will treat governance not as a brake, but as an engine of trust and resilience, building cultures where ethical use is built in, not bolted on,” Mauzaize adds.
Agentic and multi-agent AI will scale, but monitoring frameworks will be needed
Experts including Steven Webb, UK chief technology officer at Capgemini, predict AI will evolve beyond copilots into autonomous, multi-agent systems capable of planning and executing complex workflows across enterprises.
“Success will hinge on observability, orchestration, accountability, and human oversight,” Webb says. “Companies that fail to govern agents properly risk runaway costs, security gaps, and ‘shadow AI’.”
With mounting pressure to boost productivity and automate end-to-end processes, specialised multi-agent systems will be better suited than single copilot use cases, the Webb notes.
In an evolving landscape, new frameworks will increasingly emerge to orchestrate and coordinate agents safely, enabling them to collaborate and hand off tasks.
“Our research suggests sectors such as financial services, retail, and telecoms are already early adopters,” Webb adds.
However, he warns that ambition must be matched with trust.
“2026 will be about building AI readiness, platforms that allow agents to coordinate safely within robust security and monitoring frameworks,” he says.
Data becomes the ultimate strategic asset in a hybrid environment
With the proliferation of AI models, competitive advantage will increasingly depend on proprietary data and how effectively organisations manage, contextualise and govern it.
Hybrid IT architecture will dominate the market as companies rebalance cloud, edge and on-premises environments to improve security, control costs, meet sovereignty requirements and support AI workloads.
In this new scenario, treating every byte of data as valuable will become the norm, Dave Mosley, chairman and chief executive at the Seagate Technology tells National Technology News.
He predicts a year defined by data becoming the ultimate creative currency, reshaping how people create, build, and run organisations.
He says: “As AI platforms put the power of data into the hands of everyone, we’ll see a creativity Renaissance where data fuels unprecedented organisational performance and accelerates innovation in every industry as well as boardrooms, factory floors, emergency rooms and laboratories.”
He adds that more video content will be created in 2026 than at any other time in history, increasing the importance of how data is curated, stored, shared and restored.
“Treat every byte of data as if it has value, because it does,” advises Mosley.
Human–AI collaboration reshapes work, skills, and leadership
According to experts, 2026 will also be a decisive year for collaboration between humans and AI.
AI will augment managers and employees alike, boosting autonomy and productivity while reshaping talent models. According to Steven Webb, chief technology officer at Capgemini, human-AI collaboration will enter a critical testing phase, with demand rising for AI-fluent workers, senior technical expertise, and cross-functional “fusion teams.”
“UK consumers increasingly value services that blend human empathy with the speed and efficiency of AI,” Webb says. “That expectation is driving new operating models where people and autonomous agents work side by side.”
He believes that businesses will focus on fundamentals, from deciding which tasks to delegate to agents to measuring agent performance and costs.
“Unlocking human–AI chemistry will be a top priority,” he explains.
Louise Newbury-Smith, head of Zoom for the UK and Ireland, adds that AI will be integrated as a core competency for employees to thrive.
“AI fluency will become a foundational skill, and employees must learn how to collaborate with AI, interpret its recommendations, and challenge outputs when needed,” she explains.
David Sargant, director of advertising operations at SBS, agrees, adding: “The winners will combine AI’s efficiency with human judgement, making personalisation feel useful rather than intrusive, and decisions confident rather than automated.”
Bonus prediction: keep an eye on quantum!
John Roese, global head of technology and head of AI, recommends keeping an eye on quantum computing, which will continue to expand and improve.
He notes that we are rapidly closing the gap between algorithms and their requirements in terms of the number of qubits needed to perform useful work, utility or quantum supremacy, and the capabilities of quantum systems.
“This year, we’ve seen Quantinium release a 90-something qubit fully error-corrected system. IBM System 2 is 120 qubits. We see line of sight to continued error-correcting qubit growth,” he explains.
At the same time, innovation in software has dramatically reduced the number of qubits needed to run certain advanced algorithms that actually do useful work, such as for optimisation problems.
Roese points out that these developments represent enormous progress compared to a year ago.
“While I don’t believe we will see a gigantic quantum breakthrough, we are on the trajectory where at some point, that connection between the qubits needed to run an advanced algorithm and the systems that provide qubits will converge, and I think we are closer than most people think” he explains.
“There’s lots of activity, which is probably not going to transform our world in 2026, but it’s also not a decade away before we start seeing early examples of quantum utility and supremacy as these two worlds converge. So, keep an eye on quantum.”
