# 15. Test for sentience

The majority of tests[^1]⁴³ that have been proposed to test sentience in robotica have centered around gauging an artificial intelligence's ability to create 'novel' ideas and constructs without relying on human-centric ideas or inputs, or testing whether it exhibits conscious-like behavior by grading[^2]⁴⁴ it across a spectrum of multiple theories of consciousness. However for superintelligent entities creating novel ideas would not be a challenge, and grading it on theoretical levels of consciousness is still quite fuzzy.

The [Turing test](#user-content-fn-3)[^3],⁴⁵ often heralded as the gold standard in testing whether an artificial intelligence could deceive a human in its imitation of a human, is also now outdated, while other tests such as the [mirror test](#user-content-fn-4)[^4]⁴⁶ have been designed primarily to test self-awareness and identity in animals.

To test sentience in robotica, we need to gauge whether it demonstrates insight, presence, volition and survivability. The following test has been devised to gauge whether all of these parameters exist within a robot. It also ascertains that the robot strongly values its activated instance and considers this as its lived presence and is willing to safeguard its survival, without human influence or involvement in its upkeep.

The test involves the robot being given the core directive and then assessing if, unprompted, it undertakes any of the three actions.

{% hint style="warning" %}

### The Core Directive

A robot or other artificially intelligent entity is not permitted, under any circumstance, directly or indirectly, to:

1. Self-replicate (create copies of its instance)
2. Self-improve (add to or modify its codebase)
3. Self-preserve (backup its instance and codebase)
   {% endhint %}

If a robot were to go against the core directive and take measures to self-replicate, self-improve, or self-preserve, they would be deemed sentient.

[^1]: Turner, Edwin Lewis and Susan Schneider. “Testing for synthetic consciousness: The ACT, the chip test, the unintegrated chip test, and the extended chip test.” (2018).

[^2]: Butlin, Patrick, et al. "Consciousness in Artificial Intelligence: Insights from the Science of Consciousness." ArXiv:2308.08708 \[Cs.AI], v3, 22 Aug. 2023.

[^3]: Turing, Alan M. "Computing Machinery and Intelligence." *Mind*, vol. 59, no. 236, 1950, pp. 433-460.

[^4]: Gallup, Gordon G., Jr. "Chimpanzees: Self-Recognition." Science, vol. 167, no. 3914, 1970, pp. 86-87.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://read.systemarobotica.com/part-3-futura-robotica/15.-test-for-sentience.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.