The AI Gold Rush Is Broken: A View from RSA Conference 2026

Mar 27, 2026 | Transhuman / Security | 0 comments

Every booth. Every pitch. Every banner. AI.

Not just AI. Now it is “Agentic AI.” Next year it will be something else entirely. There is a flavor for everyone, and if you ask me, most of them have completely missed the point.

The Industry Is Forcing AI Where It Does Not Belong

What I saw at RSAC was not innovation. It was desperation.

We are forcing AI into every conceivable product category whether it makes sense or not. The result is a marketplace flooded with tools that look impressive in demos but struggle to deliver real value in production.

The numbers back this up. A 2025 study from MIT Sloan Management Review found that up to 95 percent of enterprise AI initiatives fail to deliver measurable business value [1]. That is not a rounding error. That is systemic failure.

Even broader adoption metrics tell the same story. Research from IBM shows that 78 percent of organizations are using AI in some form, yet only about 30 percent report significant ROI [2].

Let that sink in.

Most organizations are deploying AI not because it works, but because they feel they have to.

The Startup Explosion With No Substance

At RSAC alone, I was approached by at least four startups, each claiming to have a revolutionary take on AI. Different branding. Different dashboards. Same conversation.

Every time, I asked the same question.

Where is the value?

Most could not answer it clearly.

This is not surprising when you zoom out. According to data from Crunchbase, over 50,000 AI startups have been founded globally in recent years [3]. Yet the failure rate for startups overall remains brutally high, with around 90 percent failing [4].

AI is not exempt from that reality. If anything, it is accelerating it.

We are seeing a surge of companies built on hype cycles rather than sustainable business models. Many of these organizations are not solving real problems. They are repackaging existing capabilities with an AI label and hoping the market will not notice.

Cybersecurity Is Not a Revenue Engine

Let us be honest about something the industry does not like to admit.

Cybersecurity is not a revenue generator.

Companies do not invest in security because it is the right thing to do. They invest because they are forced to. Regulatory compliance. Audit requirements. Risk mitigation.

It is a checkbox.

That reality creates a dangerous dynamic. When you combine compliance driven spending with AI hype, you get tools that are purchased not because they are effective, but because they sound necessary.

This is exactly the environment where bad technology thrives.

This Feels Exactly Like the Dot Com Bubble

I have seen this before.

This entire AI surge feels eerily similar to the Dot-com bubble.

Back then, companies added “.com” to their name and valuations skyrocketed overnight. Investors poured money into businesses with no clear path to profitability.

Sound familiar?

Today, we are watching companies add “AI” to their pitch decks and experiencing the same irrational enthusiasm.

History tells us how this ends.

The dot com bubble burst. Hard. Billions of dollars disappeared. Companies vanished overnight. Only a handful of real value creators survived and went on to define the next era of technology.

I am not predicting if the AI correction happens next year or five years from now. But I am telling you this.

It is coming.

The Real Future of AI

Do not misunderstand me. AI is not the problem.

The problem is how we are using it.

Real AI value will come from:

  • Solving clearly defined problems
  • Augmenting human decision making, not replacing it blindly
  • Integrating into workflows where it delivers measurable outcomes
  • Reducing cost, time, or risk in a meaningful way

Everything else is noise.

Final Thoughts

Walking through RSAC 2026, I did not see a mature industry. I saw a gold rush.

And in every gold rush, most people do not find gold.

They sell shovels.

Right now, AI is the shovel.

And a lot of people are about to learn that digging without purpose does not guarantee discovery.

References

[1] MIT Sloan Management Review
https://sloanreview.mit.edu/article/why-ai-projects-fail/

[2] IBM Global AI Adoption Index 2024
https://www.ibm.com/reports/ai-adoption-index

[3] Crunchbase AI Startup Data
https://news.crunchbase.com/ai-robotics/artificial-intelligence-startups-funding/

[4] Startup Failure Statistics
https://www.failory.com/blog/startup-failure-rate