Does enterprise AI actually pay off?
The J-curve against the 95%, and the number that would settle it.
Explore section
Open questions: where we argue against ourselves
One view does not help anyone. Each set-piece here takes one open question about AI value and argues it from both sides, with no straw men, and a synthesis that does not always land in the house's favour. If you disagree with our thesis, this is your part of the site, and you are welcome in it.
The question, the optimist's strongest case, the sceptic's strongest case, what would settle it, the stakes either way, and an honest position with the uncertainty left in. Below the debates, four essays look at where the field may go next.
The debates
Is attention the new scarce resource?
Cheap generation, finite judgement. Or is the bottleneck somewhere else?
Explore section
Should AI govern itself?
Machine-speed self-optimisation versus the fox guarding the henhouse.
Explore section
Is decentralisation good or bad?
Liberation from the chokepoint, or the end of governance?
Explore section
Does value optimisation serve people or extract from them?
The instalment that argues with this site most directly.
Explore section
Is value even the right north star?
The closing question: whether the premise of this hub is the right one.
Explore section
Industry Perspectives
What optimists, sceptics and practitioners are actually saying, sourced from published research and public commentary.
Explore section
From the frontier
Emerging practice and what may come next. More speculative than the rest of the site, and labelled that way. All frontier essays
Beyond the chatbot
The six-level form-factor spectrum, and why attribution breaks as the human leaves the loop.
Explore section
Teaching AI to manage value
As AI builds AI, the objective we embed becomes the objective that scales. Read with the debate on whether AI should govern itself.
Explore section
From cloud to laptop
The decentralisation arc and the disappearing meter. Read with the debate on whether decentralisation is good.
Explore section
The spell-checker for thinking
An essay on what individuals give up when AI does the reasoning, and what personal accountability for AI use could look like.
Explore section