Branding & Strategy — April 2026

Open Midjourney. Type "modern minimalist logo for a tech company, blue gradient." You'll get four options in 30 seconds. They'll look professional. They'll look polished. And they'll look exactly like the logos generated by the other 50,000 people who typed a similar prompt this week. This is the fundamental paradox of AI in branding: the easier it becomes to produce visual assets, the harder it becomes to create something distinctive. Brand identity has never been about making things look nice. Any tool can do that now. Brand identity is about meaning — why this colour and not that one, why this typeface communicates the right personality, why the logo works at 16 pixels on a favicon and 16 feet on a billboard. These decisions require understanding the business, the market, the customer, and the competitive landscape. AI doesn't have that context. It has training data. The brands that will win in the AI era are the ones that invest in strategic thinking before visual execution. A brand strategy document — covering positioning, voice, values, audience psychographics, and competitive differentiation — is the foundation that AI cannot generate, because it requires human judgment about what matters. Every visual decision flows from this strategy. Without it, you're decorating. With it, you're communicating. Here's what we're seeing in the market. Businesses that used AI to generate their brand assets are already hitting a wall. Their logo looks like everyone else's because it was trained on the same dataset of existing logos. Their colour palette follows the same AI-optimised trends. Their brand voice sounds like ChatGPT because it is ChatGPT. When every competitor looks and sounds the same, the brand effectively disappears. Human designers bring something AI fundamentally lacks: intentional imperfection. The best logos have a quirk, an unexpected element, a deliberate rule-break that makes them memorable. The Nike swoosh isn't symmetrical. The FedEx arrow is hidden. The Apple logo has a bite taken out of it. These design decisions came from human intuition and creative risk-taking, not from statistical pattern matching across millions of existing logos. We use AI as a tool in our brand identity process, but not for the reasons you'd expect. We use it for rapid mood board generation during the discovery phase — not to create final assets, but to quickly explore visual territories and align with clients on direction. We use it for competitive analysis, processing hundreds of competitor brands to identify visual white space in the market. We use it for variation generation, taking a human-designed concept and exploring subtle modifications in colour, proportion, and layout. The creative direction is always human. The heavy lifting gets accelerated. Typography is one area where AI consistently fails. Choosing typefaces is a nuanced craft that considers readability at different sizes, cultural associations, historical context, and how letterforms interact with each other. AI will recommend "clean sans-serif fonts" because that's what's statistically popular. A human designer will choose a specific typeface because the way its lowercase 'g' curves echoes the rounded forms in the logo, creating visual cohesion that the user feels but can't articulate. The cost of getting brand identity wrong has also increased. In 2020, a generic brand was invisible. In 2026, a generic brand actively hurts you, because consumers have become attuned to AI-generated content and they associate it with cheapness, laziness, and inauthenticity. A handcrafted brand identity signals that you care about your business enough to invest in it properly. That signal matters more now than ever. Our advice to businesses: use AI for productivity, not for identity. Let it write your first draft, generate your social media variations, resize your assets for different platforms. But your core brand — your logo, your colour system, your typography, your voice — should be created by a human who understands your business and has the craft to translate that understanding into design. The AI can assist. It shouldn't lead.