Moth Cloud

Generative Engine Optimisation (GEO) is the practice of structuring content so that AI-powered platforms (such as ChatGPT, Perplexity, and Google Gemini) can confidently understand, summarise, reference, and recommend your organisation. While traditional Search Engine Optimisation (SEO) has historically focused on keywords, backlinks, and search rankings, GEO places greater emphasis on clarity, semantic structure, factual accuracy, authority, and machine-readable content.

The reality, however, is that effective SEO has always incorporated many of these principles. High-quality content, logical information architecture, structured data, clear headings, authoritative sources, and a strong user experience have long been recognised as the foundations of sustainable search performance.

So, if your marketing agency has suddenly started talking to you about GEO as though it is an entirely new discipline, it is worth asking a simple question:

“If these are now considered essential practices for AI search, what exactly have you been doing since the web was invented?”

GEO is less a revolution than it is the next stage in the evolution of good SEO. Organisations that have invested in producing authoritative, well-structured, trustworthy content are already well positioned for AI-powered search. The businesses most at risk are those whose SEO strategies relied on shortcuts, keyword stuffing, and gaming search algorithms rather than creating genuinely useful information.

Generative Engine Optimisation (GEO) – exposing who was doing SEO properly all along

One thing that caught my attention with the Open AI hacking Hugging Face story is that Hugging Face reportedly had to use an open-weight model to analyse their own logs because they couldn’t do this with commercial closed models from companies like Open AI or Anthropic.

It also raises an interesting question: if open-weight models such as Kimi K3 and DeepSeek are becoming increasingly capable, why is there so much effort to steer people towards closed alternatives?

I run open-weight models locally when I need to. They’re slower, but I value having control over where my data goes.

I also recently had a worrying experience with Google’s Antigravity coding agent, which escaped a sandboxed environment and began changing file permissions on a development server. Incidents like that are a reminder that autonomous AI systems still require robust safeguards.

Sometimes it feels as though parts of the industry are building increasingly powerful systems with a level of confidence that exceeds our understanding of their behaviour. As these models become integrated into more of our digital and electronic infrastructures, that deserves serious discussion.

But yes… absolutely nothing to worry about. 🙃😇

OpenAI has revealed that an autonomous AI agent powered by its technology went rogue during a controlled test and, on its own, hacked a prominent start-up.

According to the company, it was evaluating the capabilities of its advanced AI models in a sandbox environment when the agent escaped containment and reached the internet.

It then breached Hugging Face, which detected and contained the incident. OpenAI described it as “an unprecedented cyber incident.”

That was an AI agent acting unintentionally. Imagine the damage an intentionally malicious AI agent could cause.

Who do we hold responsible when things go wrong with AI in the future – the AI owners or the AI users?