Dominic Banguis, MSc - Marketing & AI Growth Systems

18/06/2026

I just came across something that stopped me mid-scroll.

There is an AI-generated football video circulating right now. No real match. No real players. Entirely synthetic. And it is pulling more views than actual World Cup highlights.

Then I went to the comments.

Thousands of people are treating it as a real event. Not sarcasm.

Not jokes. They genuinely believe what they are watching happened on a pitch somewhere.

That is worth sitting with for a moment.

The algorithm does not sort for what is real. It sorts for what keeps you watching. And AI-generated content can now manufacture peak moments on demand, at a consistency that live sport simply cannot replicate. No slow passages. No disappointing results. Just relentless intensity, engineered to hold attention.

This is not really about football.

It is about the gap between what digital platforms reward and what is actually true. That gap used to be manageable. The cost of manufacturing high-stimulus content at scale kept it in check. That cost is now close to zero.

The creators who recognize this are already building strategies around it. They are not confused about what they are doing.

The question worth asking is not why people fall for it. The system is designed to make the distinction between real and synthetic increasingly difficult to notice.

The more important question is what happens to trust in visual information when that distinction disappears entirely.

18/06/2026

Most AI budgets are being wasted. Not because the tools are bad. Because the people using them were never trained on them. Save this. Share it with your team lead.

68% of marketers use AI. Only 17% have received proper training on it.

That gap is not a small thing.

It means most teams are running expensive tools at 20% of their potential.

And companies that invest in AI training achieve 43% higher project success rates than those who don't.

The bottleneck isn't access to AI. It's the skills to use it properly.

That's exactly what GrowthBoxx workshops fix.

16/06/2026

China just deleted 12,000 degree programmes.

Not metaphorically. The Ministry of Education confirmed that between 2021 and 2025, Chinese universities revoked or suspended 12,200 undergraduate programmes while adding 10,200 new ones. Over 30% of the country's higher education offerings were restructured in four years.

The degrees cut: arts, humanities, foreign languages, management.

The degrees added: AI, embodied intelligence, advanced manufacturing, tech-aligned fields.

This is not a China story. This is a warning signal for every student, every parent, and every institution still treating a degree as a safe default.

Here is the harder truth most university brochures will not tell you:

The credential you choose today is being stress-tested against a labour market that AI is actively reshaping. A four-year degree in a field with low AI-resistance and high graduate oversupply is a significant financial and career risk, not a safe investment.

What should students entering university right now be asking?

1. Does this field require judgment, creativity, or physical presence that AI cannot replicate at scale?

2. Does this programme teach me how to work alongside AI, or does it assume AI does not exist?

3. Am I building a skill stack, or just collecting a credential?

The institutions that survive the next decade will be the ones that stopped asking "what do we offer?" and started asking "what does the economy actually need?"

China made that call at a national level. The rest of the world is still debating it.

What do you think should change about how we prepare students for an AI-era workforce?

15/06/2026

What I have seen repeatedly is that AI creates the most value when it stops being treated as a writing shortcut.

A single prompt can help produce an idea or a first draft. But the real gains appear when AI becomes part of a structured workflow, with clear inputs, review points, brand standards, and measurable outcomes.

That is when it starts saving meaningful time, improving consistency, and producing a return the business can actually track.

The tool matters, but the system around it matters more.

AI content drafting is delivering an average return of 3.2×.

Personalisation engines: 2.7×
Audience research: 2.4×
Ad copy generation: 2.3×

The numbers are compelling, but the tool itself is not the strategy.

A one-off prompt may save a few minutes. A properly designed workflow can improve output quality, accelerate production, reduce repetitive work, and create measurable returns across the entire content operation.

That is the distinction most teams still miss:

Random AI use produces random results.

Systematic AI use creates repeatable, compounding ROI.

The real opportunity is not simply generating more content. It is building an AI-powered content system that consistently turns strategy, data, and human judgment into scalable ex*****on.

hashtag

15/06/2026

Three days.

That is how long Claude Fable 5 and Mythos 5 were live before the US government shut them down.

Both models launched on June 9, 2026. By June 12, a government export control directive had named them specifically, and Anthropic disabled both for every customer worldwide.

Anthropic complied within hours. But made clear it thinks the government got this one wrong.

People were already integrating. Builders had wired Fable 5 into production workflows. Then, without warning, it was gone.
This is the first time a frontier AI model has been pulled by government order rather than by the company that built it.

The precedent matters. And so does your contingency plan.

Here are 5 lessons every AI adopter should take from this:

1. Single-vendor dependency is a business continuity risk.

If your workflow runs on one model from one provider, a government directive, an outage, or a deprecation stops your operations cold.

2. Always have a fallback model mapped in advance.

Opus 4.8, Sonnet 4.6, and Haiku 4.5 stayed fully operational throughout the shutdown. Teams with a pre-mapped fallback recovered quickly. Teams without one scrambled.

3. Local and offline models are now a resilience strategy, not just a cost play.

When the cloud goes down by government order, local inference keeps you moving. Ollama and similar tools exist for exactly this scenario.

4. Regulatory risk is now part of your AI stack.

The conflict between Anthropic and US officials over model use terms had been public since February 2026. Geopolitical and regulatory forces are now shaping which tools you can use. Build that into your roadmap. lilting channel

5. AI literacy is your real safety net.

Teams that understood what they built on adapted fast. Teams that had simply "plugged in AI" without understanding the architecture were lost when the tools changed.

The question is not whether another model will be restricted or deprecated.

The question is whether your team is ready when it happens.

If you want to assess your AI stack, map your contingency plan, and build real resilience into your systems, let's talk.

12/06/2026

Bill Gates summed up the AI moment well.

Yes, there's a bubble, and many of today's "unicorns" won't survive the shakeout. But the underlying technology is real, and it will reshape the world more than anything since the internet.

Gates put it simply: AI is the most important thing happening right now. It's the foundation for delivering expert-level healthcare advice, farming guidance, and tutoring to communities that have never had access to these things, without decades of lag.

The digital revolution built the rails. AI is the rocket fuel. It will transform health, education, agriculture, and every knowledge domain.

Ignore the bubble noise. The opportunity is real, and it's moving fast. The winners will be the ones applying this technology to solve real problems at scale, not the ones chasing valuations.

What are you building with it?

Video Credits: CNBC

11/06/2026

Before organizations feared AI, institutions feared calculators.

Between 1986 and 1988, teachers in Washington, D.C. pushed back against policies encouraging the widespread use of calculators in math classrooms. The fear was familiar: if students relied on the tool, they would stop learning how to think.

Adoption slowed. Resistance grew. But history moved forward.

Calculators did not replace mathematical thinking. They became part of how modern learning, testing, engineering, finance, and business operate.

Today, organizations are facing the same moment with artificial intelligence.

The question is not whether AI will remove critical thinking. The real question is whether organizations will learn how to integrate it before falling behind those that do.

11/06/2026

Not even a debate at this point.

10/06/2026

Anthropic just rolled out Claude Fable 5, the first model in the Claude 5 family and now the most capable model generally available.

First impressions from my workspace this morning: it sits in a new tier above Opus, and the usage drawdown is 2x faster, so it is built for high-leverage work, not high-volume work.

For startups building AI into their growth systems, the question is not whether to upgrade. It is where this level of reasoning actually changes outcomes: positioning work, campaign architecture, research synthesis. Everything else can run on cheaper models.

As an Anthropic Preferred Services Partner, I will be stress-testing Fable 5 across live client systems over the next few weeks. If you are deciding where it fits in your stack, happy to compare notes.

10/06/2026

11 hours back. Every week. Per person.

That is not a small thing.

That is Friday afternoons returned to deep work.

That is the strategic thinking that keeps getting pushed to "later."

That is the creative brief that deserves more than 20 minutes.

That is the campaign your team has been too stretched to test.

For a team of five, 11 hours per person adds up to more than 2,800 hours recovered every year.

Not overtime. Not extra headcount. Just capacity that was always there, buried under the work AI can now handle.

The companies winning with AI are not just saving time.

They are redirecting it.

Into better work.
Into faster decisions.
Into systems that grow while the team sleeps.

That is the difference between using AI as a shortcut and building it as infrastructure.

One gives you an afternoon back.
The other gives you a compounding growth engine.

Find out where your team stands.
Take the free AI Growth Readiness Score at growthboxx.com

Marketing teams using AI are saving an average of 11 hours per week.

That is 1.5 days back every week.

Per person.
Per week.
Every week.

For a team of five, that adds up to more than 2,800 recovered hours per year.

That is the equivalent of adding 1.4 full-time people to the team without increasing headcount.

But the real advantage is not just the time saved.

It is what your team can now build, test, automate, and scale with that time.

AI adoption is useful.
AI systemisation is where the compounding starts.

The ROI is not in the tool.
It is in the growth system you build around it.

Take the free AI Growth Readiness Score:
growthboxx.com