Progress Software’s Chief AI Officer explains why agentic RAG and compact models will define success as organisations shift from AI pilots to scalable platforms.
What’s happening: Organisations are transitioning from AI experimentation to scalable deployment, focusing on orchestration, governance and auditability.
Why this matters: The shift toward disciplined implementation with built-in governance could determine which organisations unlock competitive advantage versus those left behind by widening productivity gaps.
If 2023 to 2025 were the years of pilots and prototypes, 2026 will be about orchestration, governance and scale.
The signal across various pieces of recent AI research is consistent: adoption is widespread, and business impact concentrates where companies redesign workflows, measure outcomes and hard-wire trust and controls into the stack. And while McKinsey reports that approximately 80 per cent of companies use generative AI, yet most still aren’t seeing material earnings contribution because scaling practices and operating models lag the hype.
Ed Kiesling, Chief AI Officer at Progress Software, has outlined five predictions that all organisations investing in AI in 2026 need to be across.
The first centres on what Kiesling calls AI plumbing. For true competitiveness, organisations need to re-evaluate their AI foundation, updating retrieval, governance and audit systems to match new regulations and threats.
“Agentic RAG is now essential: while classic RAG used trusted data, agentic RAG adds multi-step reasoning, tool use and secure coordination. This offers ready-made scaffolding, especially valuable for mid-market and SMBs that lack resources for custom solutions,” says Kiesling.
“Instead of building complex systems from scratch, teams can adopt platforms delivering secure retrieval, reasoning and auditability democratising AI and reducing risk.”
Productivity gap widens
Kiesling’s second prediction focuses on what he describes as the compounding curve. The gap in productivity between the innovators, early adopters and the laggards will quickly expand. The innovators and early adopters will accelerate their ability to leverage AI to automate and orchestrate more complex tasks.
“Organisations will need to come up with innovative ways to foster growth and learning to bring the rest of the organisation along. Companies that prioritise structured learning and platform adoption will be best positioned to narrow the gap and fully realise AI’s transformative potential,” he says.
Smaller models win
While the trend in AI has often focused on building ever-larger and more powerful models, organisations in 2026 will increasingly recognise the benefits of smaller, more specialised models, according to Kiesling.
These compact models can be trained or fine-tuned on a company’s own data, making them highly relevant to specific business needs and internal processes. Because they require less computational power and storage, smaller models are less expensive to run and can be deployed in a wider range of environments.
“This allows for faster, more secure and more private AI operations, as sensitive data can remain within the organisation’s own infrastructure rather than being sent to the cloud or third-party providers,” says Kiesling.
Trust over hype
As AI hype fades, what endures is trust, Kiesling predicts. Early enthusiasm around AI everywhere is giving way to a more mature set of expectations. Organisations will demand AI systems that are transparent, auditable and fair, especially in regulated or sensitive industries.
“For example, consider a financial services firm that must comply with strict regulatory requirements. Instead of deploying a black-box AI model for loan approvals, the company implements an AI platform that provides clear explanations for each decision, maintains a complete audit trail and allows internal and external reviewers to trace every step in the process,” he says.
Kiesling’s final prediction centres on integration. The organisations that excel ahead over the next 12 to 18 months won’t be the ones with the largest models. It will come down to those who can unlock the value of the unstructured data within their organisation to drive more meaningful business outcomes for their customers.
“The secret sauce is turning scattered protocols, tribal knowledge and unstructured documentation into actionable retrieval pipelines. Agentic RAG can power this transition,” says Kiesling.
“Take for example a security team who today may rely on a slow, manual search for incident-response playbooks, regulatory requirements or architectural documentation. Instead, this same team can instantly surface the tools they need in real-time, relying on a unified, contextual retrieval layer that becomes a competitive advantage. Combining retrieval, reasoning and secure orchestration revolutionises the way teams access internal knowledge.”
Organisations aiming to succeed with AI in 2026 must adopt a more disciplined and strategic approach to building and deploying their solutions. This year marks a pivotal shift from experimenting with pilots to implementing scalable platforms, where the focus is on orchestrating and governing AI at scale.
“Agentic RAG plays a crucial role in this transformation by merging generative reasoning with robust, accountable retrieval. Every response is grounded in verifiable data, every process is clearly tracked with an audit trail and every interaction maintains strict separation between environments ensuring governance is never compromised,” says Kiesling.
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
