The Human Influence Observatory

A public observatory tracking how much influence humans still hold over the systems that run our lives.

Project Details

The project builds a public observatory for measuring and monitoring human influence across AI driven societal systems, deliberately tracking human influence and not AI capability. Building on literature by Kulveit et al. (2025), the aim is an indicator framework, a set of measurable proxies for human influence, and a platform where they can be monitored over time.

The observatory will initially focus on domains identified in gradual disempowerment literature where measurement is tractable: economic systems (human influence over labour markets, capital allocation and corporate decision-making), political institutions (human contestability, oversight and participation in governance), and cultural systems (human participation in cultural production and influence over cultural evolution).

Human influence here means whether people remain able to intervene, contest, and redirect: whether decisions can still be challenged and the objectives behind them shaped, whether increasingly agentic activity remains steerable, and how far institutions still depend on human workers, voters, consumers and participants. Because influence can't be observed directly, the observatory is built from proxies expected to track it, read over time, with attention paid to the interaction effects between domains, the early-warning indicators for feedback loops between economic, political, and cultural influence that drive GD, and the points at which human influence becomes critically compromised. I expect building that to take more than one grant, but this covers its first stage: the economic domain. It is where the measurement task is most tractable, where usable data already exists, and where the influence that propagates outward into culture and politics originates.

The observatory tracks four indicators. AI's share of GDP, considered as a category of its own rather than folded into capital so that output directed by systems can be distinguished from output directed by people. The fraction of consequential corporate decisions made by AI rather than by humans, measured through procurement data and survey data. And the volume of AI spending that no person has approved, as well as the divergence in wealth between industries that still depend on human workers and those that no longer do. The project starts with the US and UK where most data is available, and extends to a few other OECD economies where labour share data permits. Figuring out which ones make up the set depends on what the data supports, and settling that is part of the project.

The grant will help towards four outputs: a method for constructing each indicator (how it is defined, what data it draws on), datasets of the proxy data (documented data behind each indicator), the public observatory as a website where indicators can be read with the underlying data and methods visible alongside them, and a published methodology, including an explanation of which indicators resist measurement and what would need to change for them to be measurable.

Theory of Impact

The core x-risk here is loss of control, but not through a power-seeking system that seizes it. It's the gradual variety: humans lose meaningful direction over the future through many small, individually reasonable transfers of power that compound until recovery is no longer possible. Nothing has to want control as ordinary optimisation and competition do the work. Economic, cultural, and political systems are currently aligned with human interests largely because they depend on humans as workers, consumers, creators, voters, and the people who run them. As AI becomes cheaper and more capable, the case for replacing humans in each of those roles strengthens, and as it does, the levers humans hold over how systems behave weaken. Oversight erodes for the same reason: the systems stop being tethered to the people overseeing them.

What makes this existential rather than merely disruptive is that it locks in. The erosion runs through locally rational, voluntary choices with no single handover that visibly crosses a line, so there is never an obvious moment to intervene, and it can very well resemble abundance on the way down. Unlike a decisive takeover, there is no adversary to coordinate against and no fix that dissolves it: it survives the alignment solution we bet on as well-aligned components can still aggregate into a system that no longer runs on human preferences. The endpoint is a world that still functions, maybe even prospers on the usual metrics, but is no longer under meaningful human direction.

People

Shalina Prakash

Team Member

Grants Received– no grants recorded

Add grant

Funding Asks

grantmaking.ai Launch Round

Applied

Minimum

$14,784

Ideal

$22,176

How the money will be spent

Minimum: $14,784

Maximum: $22,176; this is detailed in the sheet below.

Shalina Prakash <> Grant Budget

I will be the only researcher on this project, and the budget is a stipend for my time. At $22/hr, the ideal buys six months full-time, whilst the minimum buys four.

At the maximum the project delivers four things:

> the indicator framework: what the observatory measures, and why those dimensions matter

> an open dataset of the proxy data, sourced and documented

> a website as a public observatory, where these indicators can be monitored over time, together with the underlying data and methodology

> a published methodology describing the conceptual framework, indicator selection, measurement approach, and limitations, enabling others to reproduce, critique, and extend the work

The minimum delivers the same indicator framework, but two tractable indicators, the observatory and the methodology, and a brief noting what constructing the others would require. Either version will result in a working observatory, with its methodology and data released publicly so the work can be extended; I expect continued expansion and regular updating to be a separate, later stage beyond the scope of this initial grant.

Discussion

No comments yet. Be the first to share your thoughts.