The AI multiplier effect, brand equity and startups over doing it

Everyone’s building faster. Want a product shipped in three weeks? Done. Lovable.ai has got you. Need a website in 10 seconds? Heyboss.ai is doing it. You want a cool brand? Canva, Adobe or any number of companies are going to do it for you by the time you’ve finished this blog…

You are going to finish reading this, right?

“Brilliant” the team says. “We're getting to market, we are testing our product and our messaging ASAP. VCs love it, the team is seeing things happen. WE’RE DOING IT BABAAAAY!”

Or at least, that’s what they think they’re doing.

Two years ago, AI was the shiny new bolt-on. Now it’s driving downward pressure from C-suite demanding teams to find a way to integrate it quickly, damned the expense and damned the quality.

Product decks are full of it. In fact, in the last 24 months of running a consultancy built for early stage companies, I have only come across two (two.) companies that do not have AI as their core product. (Spoiler - yes, they are finding fundraising difficult.)

Teams are stacking tools like they're free snacks and investors are nodding along. Leadership teams are still quizzically saying to teams across the organisation, “We simply MUST get some more AI in here, for now”.

What’s happening when with the day-to-day operations, though, is that AI doesn’t fix bad thinking… it multiplies it. 

If your inputs are vague, bloated, or just plain wrong, AI won’t correct them. It’ll give you 100 variations of the same half-baked idea. Faster, louder, and with all the confidence of a tool that’s never had to face a customer. 

That’s not acceleration. That’s… trash?

And like the dawn of the SaaS tool revolution, where we’re using so many, and being so productive, you’re missing the details that make you look… 

Stupid.

Today, AI adoption pressure is downward in search of efficiency, not customer up in search of value.

Right now, companies are layering on tools like Notion AI, ChatGPT, Midjourney, Claude, Gemini, a thousand plugins and SaaS wrappers, but the pressure testing is only coming from up the chain, where leadership teams are forcing AI tools onto teams, and the teams are forcing them onto the end user despite backlash and those users saying “I didn’t ask for this.”

We’re already seeing this in major tech companies - where what used to be the “bottom right app menu” has now become the AI prompt button, forcing users to use AI by using their muscle memory against them.

Two extremely familiar examples, for all, will be Spotify and Amazon. Both heavily invested in their own proprietary tools have forced their way into centre stage for the user.

Here's Spotify, pre & post AI launch:

Amazon, pre & post AI launch:

Not to mention Meta and their forced integration of AI (without a meta/facebook account at all), or even Reddit’s ‘answers’ section, (it’s finally given them an excuse to fix their search function, at least)

OK, that’s four major tech brands, what does this have to do with me and my startup/company…

If you’re trying to change behaviour, change a category, or transform the way something happens, forcing AI onto your users isn’t going to win you business. 

If you are trying to build the next big AI tool in your space - you need to make sure that you are absolutely bloody brilliant at already doing the thing without AI. Otherwise, no matter how powerful your tool is, the key stakeholders involved in buying your product won’t care, you’ll be too risky, too short-sighted, and too uninformed.

By all means stack as many tools, shortcuts and efficiency blasting build-your-way-brilliance-by-breaking-everything, but be prepared for what happens next.

Several things:

1. The vibe check is coming.
   Customers, clients, readers — real people — can feel when something’s off (witness the em dashes, WITNESS THEM!). When it reads like a bot. When it sounds like everyone else. When it doesn’t say anything new. Your brand becomes just another soup of AI-scented sameness, and blending in won’t be enough for you to win
   
   

2. The costs will catch up.
   When the VC money dries up (and it will), so will the tolerance for stacking ten tools on a $0 revenue model. Just like we realised Uber wasn’t really cheaper, or that Deliveroo’s “free delivery” wasn’t. The AI subscription boom will collapse back into realism. And you’ll need a strategy that doesn’t vanish with your runway.
   
   

3. Actual people hate it.
   Sure, your product manager loves using it, your dev team love shortcutting code snippets, marketing loves stitching bullet points into blog posts without actually having an A-level in English, but the moment your AI slop hits the inboxes of your hard-earned customers, disengagement is instant
   
   

4. AI brain drain will happen.
   As the skill set involved in AI becomes more and more intrinsic to some, those without the tool knowledge will become even more separated from the organisation. The risk this presents is that when those experts leave, replacing that expertise becomes doubly as difficult - the team left behind won’t know how something works, they won’t know why, and they might not even be able to diagnose the systems themselves in the event of change
   
   

5. Critical dependency on startups that might fail will cause you to fail, too.
   As hundreds of AI companies in the same category duke it out for dominance and to be named the “best practice” tool of their type, those that haven’t got lucky or have been cut loose by the VCs backing them will shutter, and they’ll take your data, and your workflows with them.
   
   

6. Regulation will catch up
   It will take time, but regulating how AI operates, where it’s allowed to reach its hungry language models into, will eventually be met with walls designed to safeguard that data, not feed the 000’s of AI companies that suddenly no longer have anything to learn from
   
   

Next, it’s the knock on effect of all these little challenges that chip away at the brand equity you’ve been painstakingly building.

Brand equity has always been hard to build and easy to lose, AI is making it even easier to lose.

The CEO of Duolingo has come off, as I hope he minds me saying, a massive tool:
https://www.linkedin.com/news/story/duolingo-ceo-walks-back-ai-memo-7350010/ 

Being so out of touch with the end user, and even your team, is a dangerous way to operate and its costing valuable team members who are exiting the organisation, and paying customers who no longer trust you.

This one of a dozen extremely public AI u-turns that are happening in 2025 by companies across tech, fashion, B2B and beyond, each determined to implement AI policy that doesn’t align to user or team behaviours without even consulting them on how it's going to happen.

The real skill? Knowing when to back off.

Here’s what’s actually useful right now: 

People who are good at their job, and good at using AI to amplify it.

Not prompt junkies. Not tool collectors. Not this viral generation of tinkerers. 

Just people who know when to go full throttle and when to write the damn thing themselves.

Because the future isn’t AI or human. It’s human work, AI-accelerated, not AI-generated with a cursory glance from an overworked marketing grad struggling to make a living under the guise of being “human-in-the-loop” AI.

This means that you do actually have to train your team and learn the basics, the mids and the advanced bit of the tech that you’re using/building - so you can understand it as intimately as your customer needs to use it and pay for it. So don’t skip out on learning, it will make your AI multiplier stronger. (There is a joke about c-c-c-combo’s in here, somewhere.)

The teams who get the balance right? They won’t just ship faster.
They’ll ship better, they’ll build brand equity and loyalty faster because they’re putting enough care and attention into their products and services that they would work brilliantly without AI doing all the thinking.

Hell, they might even stumble upon an authentic relationship with other human beings giving them money, wouldn’t that be a lovely thing to happen?