AI is increasingly being used across industries and functions, and marketing is no exception to this. Marketers across the globe are currently trying to integrate AI into their creative workflows, day-to-day operations and team structures.
Despite the depth of capabilities in the latest multi-modal models such as GPT5, these chatbots are still not very creative, and not as smart as a human. Nor do they have the ability to really think for themselves or act as an independent unit or ‘employee’ (yet).
Furthermore, once we move past the hype around LLMs and chatbots such as ChatGPT, we realise that AI actually encompasses a wide spectrum of Machine Learning models including simple statistical models that can be built internally, complex algorithms built by companies such as Meta, and more complex tools that can sit on top of LLMs or take advantage of multiple models.
As this revolution unfolds in front of our eyes, we can begin to realise that true productive use of AI may lie not in replacing humans, but in enabling humans to become more productive, do more, and build more.
Real use cases
But how can AI actually enable marketing to make a step-change in the way we operate, without becoming reliant on AI ‘slop’?
How can we lean into AI tools to enable our teams to do more with less, and to increase efficiency, visibility, cohesion and results?
The answer in 2026 may lie in 4 key trends which have emerged for me in my role as a marketing leader in Stakemate. We use all of the tools below on a daily basis, and have plans to expand our usage of them in 2026.
1. Using AI to scale brand-safe creative
Tech companies, especially startups, are likely to face constraints on time, cost, and visual resources for creative production, yet need brand-consistent content at scale. Whether for OOH campaigns, TV commercials, paid media ads or email headers, AI puts high quality imagery (and video) within reach for small and medium tech companies without having to invest in photography or expensive creative production.
At Stakemate, we have developed internal tools that use image-generation models trained on brand-safe guidelines to create high-quality, consistent creatives across different campaigns. We have even been able to create AI ‘actors’ which can appear in different scenarios, outfits and activities across our customer lifecycle.
My team and I built this tool from the ground up using AI image-generation models, with three key features:
- Brand safety: As a regulated business, individuals appearing in our marketing need to look over the age of 25. By creating AI ‘actors’ that appear consistently across ads, we have lowered the compliance overhead.
- Speed: These models produce creatives within seconds rather than days. From briefing to final production of the finished asset, we now have a production time of just a few hours.
- Quality: The produced creatives consistently meet a quality that would not be otherwise possible without real-life photography. As a challenger brand, we are able to signal quality and capability to our customers without the expense of real photography.
Our experience at Stakemate shows this approach is feasible, scalable and incredibly efficient and effective, and we can expect more and more companies to adopt AI creative production in 2026.
2. Leaning into AI to create internal tools
Consumer tech companies have various touch points with customers, and have different teams communicating with them, (e.g; customer service and CRM). A common challenge for startups and scaleups in this space is keeping communications consistent with brand guidelines.
AI–when used properly–seems to offer a good solution to this problem. The difficulty is in creating a strong, consistent tool that can be used by all team members without too much training.
The solution I built at Stakemate (using AI ‘vibe coding’ tools) is called Charlie.ai.
- Charlie is a web app with a user-friendly UI.
- Team members can input what they want to say, include customer queries where relevant, pick a scenario and a tone, and generate consistent, on-brand content every time.
- Uses multiple levels of training including past outputs to enable our team to confidently create brand-safe communications.
- Is now being used thousands of times every month by our team members.
We expect to build additional functionality into Charlie.ai, and will be experimenting with building further AI tools such as analytics reporting or brand production tools for internal use in 2026.
The rise of ‘vibe coding’ platforms such as Lovable represents an incredible opportunity for entrepreneurs. But the real benefit may be to non-technical teams within businesses who need simple internal tools that they could not otherwise build themselves. We also used AI to build and launch our Stakemate Yearly Wrapped (VAR) project.
Expect marketing leaders to do more in this space in 2026 as it may be the year marketers break out of relying on other teams to build the tools they need and use vibe coding to build it themselves.
3. Trusting algorithms and enabling them with signal engineering and creative
In the past few years, old-school marketers reliant on detailed targeting criteria in ad platforms have been increasingly sidelined. The rise of AI tools such as Meta’s Advantage+ means algorithms are more effective at finding customers for a product than marketers defining an ICP using detailed targeting.
New developments such as the launch of Meta’s Andromeda algorithm will only push this trend further, and ad buyers’ job is no longer telling the algorithm what to do.
Instead, effective ad buyers in 2026 will:
- Work to tell the algorithm what a good customer looks like, including by sending it conversion data within the conversion window.
- Continuously produce strong ad creatives.
Andromeda especially will enable the best companies that produce a large number of Meta-native ad creative (think UGC, podcast, talking head videos and reels) to outperform their peers and competitors, making creative production more important than ever.
In 2026, every business will increasingly be a content and data business, and the outliers in performance will be those who are able to outshine competitors on both.
4. Using AI models to optimise targeting and lifecycle marketing
Finally, as more businesses adopt hyper-personalisation and use CRM tools such as Braze or Customer.io (we use the former), using AI throughout the customer lifecycle becomes more and more important.
At Stakemate, I have been leading a team of engineers building AI models that classify our users into different segments, and help us personalise our lifecycle marketing to each individual customer.
Personalisation is no longer limited to adding a user’s first name into the email, but rather involves a mix of:
- Personalising offer being communicated
- Communication timing
- Communication channel
- Message content
Using AI models we have built internally, as well as by using those that exist on tools such as Braze, we are able to ensure our messaging is relevant to our customers, leading to retention and conversion metrics that far outperform our peers within the industry.
In 2026, effective marketing teams will use AI models to classify their users into different interest and engagement groups and optimise their lifecycle marketing for each individual user.
They will also use tools such as Movable Ink and Liquid to show customers different content depending on when and where they are opening emails or other messages. An effective lifecycle marketing strategy in 2026 will involve messaging that always feels ‘right’ to customers and never feels like it was a ‘blast send’.
Looking forward
If 2025 was the year of AI ‘slop’, expect 2026 to be a year where industries slowly adopt more realistic use cases for this groundbreaking new set of technologies. Marketing operators need to be proactive in their adoption of AI tools, and not wait for other teams such as engineering to bring these tools into the company.
The outstanding marketers of 2026 will use AI across different work streams, integrate it into their tools, and learn to trust the algorithms as partners. This can be achieved by seeing AI as an expansive set of tools and products that solve problems on varying levels, and not just the interface of ChatGPT.