Businesses have grappled with what GenAI and, subsequently, Agentic AI will mean for enterprise workflows and processes. But with a greater understanding of AI use cases for driving growth, alongside the implementation challenges involved, the focus has now turned to value and deliverable outcomes.

According to GlobalData Thematic Intelligence’s Tech Predictions 2026, the Agentic AI ecosystem will continue to expand rapidly “despite uncertainty about returns”. The research and analysis company expects the global AI market to reach $642bn in 2029, up from $131bn in 2024, supported by a growing ecosystem of integrators, startups, and tech giants.

Verdict asked a cross section of senior technology professionals about their enterprise AI predictions for the coming year:

Isabel Al-Dhahir, principal analyst at GlobalData

“In 2026, dozens of companies will enter the market to provide industry-specific agentic AI solutions. Despite this momentum, enterprise adoption in the coming year will continue to be held back by uncertainty about whether these tools can add demonstrable business value. Agentic AI is capable of delivering far more value than existing GenAI tools due to its greater autonomy and methodical approach to reasoning, problem-solving, and decision-making. The next step is crafting these agents for high-value use cases that merit substantial long-term investment.”

Chintan Patel, EMEA CTO at Cisco

“2025 was the year UK enterprises learned what it really takes to implement agentic AI. The biggest lesson was that agents can only operate effectively when built on unified data, strong governance and an integrated ecosystem. Many organisations discovered that DIY approaches, from training in-house models to stitching agents onto scattered data sources, created more risk than value. Where adoption succeeded, businesses paired enterprise-grade platforms with clear guardrails, allowing agents to combine predictive insight, generative capability and autonomous action across systems.

“In 2026, we’ll see a shift from experimentation to scale. Agents will begin acting as true digital co-workers, managing complex workflows end-to-end rather than automating isolated tasks. The differentiator won’t be who has the most sophisticated model, but who has the most coherent data strategy and governance framework. The UK enterprises that pull ahead will be those that move past ‘automation dressed up as AI’ and invest in agentic systems built for trust, interoperability and continuous oversight.”

Rajeev Singh CEO at Smartsheet

“The question isn’t: How many AI agents do you have? It’s what have they actually improved? We’re moving from AI as a novelty to AI as a measurable business driver. The companies that win will be the ones who treat intelligence, human and artificial, as a system for real outcomes: better speed, quality, and margins. Outcome-first AI will end the pilot purgatory.”

Christian Pedersen, chief innovation officer at IFS

“While many companies have focused on AI-first strategies, the next phase is about seamless integration: embedding AI into existing systems and workflows. Integration is moving beyond the digital and into the physical. We’re now seeing the emergence of robotic AI deployed to inspect assets, gather real-time operational data from the field, and feed that information directly into agentic AI systems that include digital workers and analyse, decide, and trigger action.

“This closes the loop between the physical and digital worlds in sectors like manufacturing, energy, and telecommunications, where AI will underpin everything from scheduling and inventory optimisation to predictive maintenance.
“This shift highlights AI’s entry into its industrial phase: embedded, standardised, and measurable. Competitive advantage will hinge on how effectively it is deployed to drive performance and resilience, not merely having it in the toolbox.”

Richard Shaw, technical GM UK&I at Databricks

“In 2026, AI agents are set to become part of the everyday rhythm of how companies operate, quietly supporting the flow of work and improving how teams coordinate and make decisions. As this shift accelerates, businesses will need much clearer guardrails that define who is responsible for an agent’s decisions, how data is used and what good looks like when humans and AI work alongside one another.

“Crucially, 2026 will also force a fundamental rethink of data architectures within organisations. The longstanding separation of operational and analytical data will become one of the biggest blockers to scaling AI. Many organisations will begin dismantling the traditional OLTP/OLAP divide and moving toward unified, modern data foundations. This shift will reduce duplication, increase reliability and give AI agents the low-latency, high-quality data they need to perform safely and autonomously.

“With these foundations in place, agents will become more capable and embedded in daily processes, taking on increasingly complex, multistep tasks across industries. Success will depend not on model type, or size but on workforces equipped to collaborate with AI through practical, domain-specific upskilling. The companies that prioritise unification and governance will operate more like connected networks, with humans and intelligent agents working in tandem.”

Louise Newbury-Smith, head of UK&I at Zoom

“2026 will be the year that Agentic AI moves from a novelty to a foundational technology, embedding itself across all touchpoints of the enterprise. Transitioning from AI assistants, this technology will take actions, coordinate tasks and execute workflows without human prompting. In more advanced scenarios, teams will have AI chain-of-command systems where agents talk to each other and reduce manual collaboration.

“In 2026, AI will fundamentally change the purpose of meetings with the emergence of AI avatars and agents. Instead of focusing on who attends, organisations will focus on what gets achieved. Agentic AI will continue to be used to summarise discussions, assign tasks and follow up automatically, meaning employees will gain back significant time normally spent preparing for, attending or recapping meetings.”

Sam Liang, CEO, Otter.ai

“Typing will feel as outdated as fax machines in 2026. AI-generated content already outnumbers human-written content on the web, and over the next decade, writing will fade into the background. However, people will never stop talking, and that’s why, as early as next year [2026], voice will become the new dominant UI for work.

“While we ideate, collaborate, debate and decide out loud, voice AI will listen, interpret, and execute in real-time. Instead of reaching for the keyboard, we’ll simply say what we need, and it will get done, with most written content created by AI from voice conversations. In an increasingly AI world, in 2026, the most powerful tool won’t be another algorithm; it’ll be the human voice that’s uniquely ours.”

Fernando Lucini, global ML engineering lead at Accenture

“AI continues to evolve at extraordinary speed, with the real value for businesses being built over the long term. We’re still in the early innings of the biggest technology transformation we’ve seen in decades – with agentic AI being one of the most exciting frontiers, because it brings a ‘task-orientated’ personality to AI that can autonomously perform tasks in a business.

“Next year [2026], organisations will want to move from pilot projects to scaled transformation – which means addressing disparate data, trusting the technology’s outputs, and fine tuning a company’s digital foundations so people can make the most of intelligence. The real advantage will lie with organisations that pair intelligent systems with their own proprietary data. Differentiation will come from quality, governance, and integration – and the ability to turn unique data into sustained competitive value.”

Etay Maor, chief security strategist at Cato Networks

“For 2026, AI will be strategic to daily workflows, networks, and decision-making systems. However, as adoption picks up speed, so will the risks. As AI grows more autonomous, security teams must replace fragmented tools with unified controls that simultaneously govern AI models, protect sensitive data, authenticate user identities, and monitor network behaviour.

“The emergence of ‘ask-and-act’ browsers (like OpenAI Atlas) will transform the web from a passive library into an active operating system, but this agency invites new forms of exploitation. As we have seen with the recent URL-masking jailbreak, threat actors can now weaponise context, tricking assistants into bypassing security checks to execute malicious commands.

“To safely integrate these tools, security leaders must urgently pivot to a defence-in-depth strategy. We must stop treating AI agents as generic tools and start managing them like employees: assigning unique identities with strict permission boundaries, classifying data at the source, and ensuring that high-risk actions always trigger a human approval workflow before an agent can act.

Nitish Mutha, co-founder and CTO of Genie AI

“For the last two years, we have treated AI as a ‘Co-pilot’: a tool that politely drafted clauses and summarised redlines under strict human supervision. By 2026, that passive era will effectively be over. We are entering the age of the active Autonomous Legal Agent.

“These agents will no longer just use tools; they will be the workflow. We will see the first generation of AI that handles end-to-end legal matters unassisted: interviewing clients to extract facts, researching business-specific context, and executing full drafts without a human keystroke.

“Most critically, the negotiation friction will vanish. We will see the emergence of ‘Agent-to-Agent’ negotiation – highly personalised to their specific business’s risk playbooks – negotiate standard agreements in minutes rather than days, flagging only true ‘deal-breakers’ for human review. This shift will fundamentally alter the law firm model, as corporate legal teams use these agents to bypass external counsel entirely for routine work, keeping the institutional knowledge and the savings in-house.”

Sven Peters, DevOps and AI expert at Atlassian

“Forward-thinking organisations will realise that rolling out AI tools and tracking usage is pointless unless those tools drive real, measurable impact. Integrating AI into everyday workflows is essential for turning potential into performance. The new gold standard will be tangible outcomes – faster innovation, better decisions, and breakthrough results.

“Open, collaborative organisations where people share data, goals, and impact metrics will avoid silos. As AI takes over routine tasks, the most valuable skills will shift from doing to directing: framing the right problems, setting smart constraints, curating context, and coaching systems to deliver what matters. Strong culture becomes the multiplier.

“The smartest leaders will redesign teams around a powerful pairing: AI-intuitive emerging talent who iterate quickly, and seasoned operators who know how to set direction, manage risk, and teach both humans and agents. Together, they’ll measure impact, not inputs.”

Mandy Andress, CISO at Elastic

“Some called 2025 the year of the agents but 2026 is when we’ll see whether those investments actually deliver. As organisations move agents from pilots into production, adoption will accelerate. In tandem, so will the security consequences.
Done well, agentic systems can strengthen defence. They can reduce noise, speed up triage, and help teams spot issues earlier by understanding how users, systems, and code normally behave. Behavioural analytics is what makes this possible. Agents that recognise subtle deviations (not just obvious indicators), give security teams a real advantage.

“But broader deployment also expands the risk surface. If you introduce agents that don’t understand baseline behaviours or fail to put the right guardrails in place, you can automate poor decisions, miss early signs of compromise, or create inconsistent enforcement. Adversaries will look for exactly these weaknesses.

“The organisations that benefit most will be the ones that treat agents like any other critical control: testing how they behave, validating what they surface, and ensuring they work alongside, not around, existing safeguards. Those that rush deployments without that discipline will heighten instability and increase the likelihood of avoidable incidents.”

David James, chief learning officer at 360Learning

“The biggest AI trend of 2026 will be the rise of the Chief learning officer (CLO). Companies are beginning to see that the real differentiator is not the technology itself but whether their people can use it to drive performance. This is a huge responsibility for the CLO. AI is disrupting skills at record speed. By 2025, almost half of workers’ core skills will shift, and we’re already seeing skills become outdated in as little as two and a half years. That makes continuous, embedded learning a business imperative.

“The CLO is the only executive who bridges strategy, capability and performance at the speed AI demands, connecting changing work with changing expectations and building the systems to support people through it. This is where AI skills mapping is an advantage. This works dynamically by alerting individuals and their line managers of changes to proficiency. Then development can be laser-focused on bridging emerging proficiency gaps. With a CLO in the C-suite, this becomes a performance engine. Without one, AI investment is at greater risk.”

Jody Gajic, director of Pearson Labs

“At the beginning of the year, I said that this wasn’t the year of agents, and qualified it as being the decade of agents. 2025 was a whirlwind of announcements and incredible development, but adoption for real value will take a few more years yet. 2026 will continue at the same blistering pace and will be the year Agentic AI shifts from a temperamental tool to a productive partner. In certain use cases, particularly coding, intelligent agents are moving from autocompleting repetitive work, to acting autonomously for longer periods, but still very much with human in the loop.

“We can expect this to expand for greater collaboration with employees in different areas of the business, solving problems, streamlining workflows, and connecting data across business functions allowing employees to focus on higher value tasks and upskill onto the next level. However, the real differentiator won’t simply be early adoption. It will be organisations that learn to manage agents as part of the workforce, embedding them responsibly with the right standards, oversight and a commitment to upskilling people to direct, supervise and improve these systems over time.”

Simon Pettit, AVP UK&I at UiPath

“Agentic AI dominated 2025, advancing from concept to deployment. In 2026, organisations will be under growing pressure to prove their agentic projects can deliver measurable results and return on investment

“While 73% believe their agentic projects will deliver next year, only those businesses targeting high-pain, high-gain areas, and redesigning workflows to fully embed AI agents, will realise that promise. Success will mean redefining ROI metrics and focusing on differentiation, not groundwork, by using prebuilt, domain-specific solutions and reusable workflows to accelerate value.

“We’ll also see broader adoption of multi-agent systems to enable smarter task coordination and unlock meaningful operational scale; 75% of organisations plan to deploy multi-agent frameworks within the next 18 months.

“2026 will be the year agentic AI proves its worth, and only those willing to approach deployment strategically, and with clear objectives in mind, will stay ahead.”