I spent nine months tracking down what happened to the eight researchers who wrote the 2017 paper that changed AI forever. What I found will make you rethink everything you’ve been told about the “democratization” of artificial intelligence.

In June 2017, these eight Google researchers published “Attention Is All You Need.” The transformer architecture they invented now powers every AI system you use, ChatGPT, Claude, Gemini, everything.

Today, all eight have left Google.

Their companies are collectively worth $15+ billion.

Illia Polosukhin, 28 at the time, now holds crypto assets worth $2.5 billion.

Noam Shazeer just got a $2.7 billion payout, the largest in AI history.

Aidan Gomez was 20 years old when he co-authored the paper. His company Cohere is now valued at $7 billion.

Meanwhile, I spoke with workers in Kibera, Kenya’s largest slum. They review the most traumatic content imaginable for $1.32 per hour, labeling data so these AI systems can be “safe” for consumers. One worker told researchers: “By Friday, you are disturbed from thinking through that picture.”

OpenAI paid the intermediary company $12.50 per hour per worker.

The workers got $2.

That’s an 84% markup on human suffering.

In Democratic Republic of Congo, 40,000 children, some as young as six, mine the cobalt that powers the AI chips. They earn $1.75 per day. When families sued Apple, Google, Microsoft, and Tesla, a US court ruled the companies weren’t liable. The relationship was just a “buyer-seller transaction.”

I tracked every dollar. Here’s what I found:

– The Global South holds 1% of AI computing power despite 85% of world population
–  Africa has 0.04% of global AI compute capacity
–  Training costs increased 4,300% since 2020, now $20-100 million per model
– By 2030, 30% of US jobs could be automated, 800 million globally
– 79% of working women face high automation risk vs 58% of men
– Data centers consumed 5 billion gallons of water in 2022, projected to double by 2028

This isn’t about technology. It’s about power.

Eight brilliant researchers made a breakthrough. Their innovation was real and important.

But what happened next, how publicly funded research became private billions, how value concentrates in Silicon Valley while costs export to Nairobi and Kinshasa, how 40,000 children dig in toxic mines so executives can discuss “responsible AI” at conferences, this is a choice.

It’s the choice we make when we don’t ask: Who benefits? Who pays? Who decides?

The research is published. The pattern is clear.

Now the question is what we do about it.

Full investigation with 10 data visualizations, 80+ cited sources, and three scenarios for the future at Aylgorith.”