It's tempting to predict we'll have fully rolled out self-driving cars by 2026, but good things take time. Here's 5 predictions the Cogworks consultants expect to see more of very soon.
This article contains links to Cogworks pages, products or services
For every article containing links back to Cogworks’ website, we donate £50 to our amazing partner, Community Tech Aid, and any successful partnership will result in another £100 donation to help fund purchases to repair and refurbish donated technology.
Predictions about technology have a habit of swinging between two extremes
On one end, we imagine a future that arrives far faster than reality ever allows. On the other, we underestimate how much quietly changes beneath the surface while no one is looking.
Recently, I spent some time talking with the Cogworks team about where digital actually feels like it’s heading over the next couple of years.
What followed wasn’t a single bold prediction, but a series of 5 subtle patterns that don't feel too far fetched...
1. Micro-automation over big AI bets
Adam Shallcross - CEO and Founder of Cogworks
Adam’s perspective highlights a shift away from large, transformative AI programmes and towards smaller, more meaningful improvements inside teams. These micro-automations don’t make headlines, but they quietly remove friction from day-to-day work.
“Micro-automations will become more common, with teams automating small, repeatable processes to improve efficiency rather than chasing large, complex AI initiatives.”
They’re easier to adopt, easier to maintain, and far more likely to stick, helping teams focus on progress rather than managing complexity. Alongside this, we can expect continued growth in the use of AI agents across businesses, but in a far more purposeful way than early experimentation suggested!
More from Adam
2. Moving past the “AI everywhere” phase
Thom - AI Architect
Thom’s view reflects a clear move away from experimentation for experimentation’s sake. While the early phase of AI adoption helped teams explore what was possible, it also introduced noise, duplication and low-value output.
“I expect organisations to invest increasingly in agent-to-agent setups with more lightweight and focused Small Language Models, as people move past the initial ‘AI everything’ phase.”
As we head towards 2026, the focus is shifting towards practical automation, systems designed to solve specific problems well, rather than showcase the latest capability. This comes with a growing awareness of the cost of over-consuming, or over-producing, AI content that adds little real value.
3. In the age of AI, software quality and trust is more important than ever
Mateusz - QA Engineer
Mateusz’s perspective grounds the conversation in an important reminder: when systems rely on the same code patterns and training data, mistakes can be replicated at scale.
“AI makes it easier to build websites and applications, but simpler creation doesn’t automatically mean higher quality.”
This makes reliability, security and performance more important, not less. As a result, demand for strong manual and automated testing is increasing, with software quality and trust becoming key differentiators for AI-driven products.
4. AI embedded into everyday workflows
Imran - Head of Development
Imran’s prediction specifically points to tools like GitHub Copilot, which are becoming increasingly embedded into developers’ everyday workflows. Rather than relying on standalone AI tools or switching context between platforms, Copilot sits directly inside the development environment.
"Copilot provides AI-powered coding assistance directly inside the development environment, making it more convenient than many standalone tools.”"
This allows developers to write, understand and refactor code as they work, with AI support available at the point of need. That convenience ( combined with improved quality and flexibility) is why tools like GitHub Copilot are likely to see wider adoption over the next couple of years!
5. Maintaining AI becomes just as important as adopting it
Kas – Head of Delivery
Kas raised one of the most important — and often overlooked — points in the discussion: what happens after AI is adopted. Many organisations are already realising their early AI tools weren’t built with long-term architecture or support in mind.
“By 2026, the focus will shift from adopting AI to maintaining it properly.”
Poorly maintained AI creates risk, inefficiency and weak user experiences. In contrast, well-supported systems deliver real value and build trust over time, making maintenance a core delivery discipline rather than an afterthought.
More from Kas
A quieter kind of progress
This year's predictions are less about ground-breaking world dominating AI and more about maturity.
Instead, it looks like teams are ready to step back, slow down and build digital products:
- that last
- with more intent
- with a clear understanding of where AI works and where it doesn't.
Progress may feel quieter than the hype cycles we’ve grown used to, but it’s likely to be far more meaningful.
Want to know how you can use tech to bring real value to your business and your users for 2026? Chat to Cogworks consultants anytime.
Innerworks and Cogworks are proud to partner with Community TechAid who aim to enable sustainable access to technology and skills needed to ensure digital inclusion for all. Any support you can give is hugely appreciated.