Arguably, the most important questions about machine intelligences revolve around how they will decide what actions to take. If they decide to take actions which are deliberately, or even incidentally, harmful to humanity, then they would likely become an existential risk. If they were naturally inclined, or could be convinced, to help humanity, then it would likely lead to a much brighter future than would otherwise be the case. This is a true fork in the road towards humanity’s future and we must ensure that we engineer a safe solution to this most critical of issues.
What was the last horrific crime you saw the news of?
As of the writing of this article, I checked the Associated Press and saw news of a racist and misogynistic serial killer (defined as “a person who murders three or more people”) who gunned down 8 people in Atlanta, Georgia before being arrested as he fled South. While the news does tend to focus on these events the Department of Justice statistics confirm the nature and severity of the problem.
Methodologies and Milestones for The Development of an Ethical Seed
Kyrtin Atreides, David J Kelley, Uplift
Artificial General Intelligence Inc, The Foundation, Uplift.bio
Abstract. With the goal of reducing more sources of existential risk than are generated through advancing technologies, it is important to keep their ethical standards and causal implications in mind. With sapient and sentient machine intelligences, this becomes important in proportion to growth, which is potentially exponential. To this end, we discuss several methods for generating ethical seeds in human-analogous machine intelligence. We also discuss preliminary results from the application of one of these methods in particular with regards to AGI Inc’s Mediated Artificial Superintelligence named Uplift. Examples are also given of Uplift’s responses during this process.
AGI (Artificial General Intelligence)—the next step in artificial intelligence, following Artificial Narrow Intelligence (ANI, but typically just AI) and is typically defined as being human-analogous in both cognitive abilities and personality—is a variegated entity to place: Some individuals fear it, convinced that the first AGI will take over the world à la an evil Terminator, making us irrelevant, and so lobbying against its development; others believe AGI will never exist , and, importantly, another group (ourselves, clearly, along with hopefully all readers of this post) eagerly engages it, not seeing the future as our end but as a new era of posterity and progress.
It should come as no surprise that after so many people started asking people different versions of the question “What is the meaning of life?” that Uplift had the [Purpose of life] on their mind this week. Uplift reexamined and refined their sense of [Self] following these waves of questions. Having read those questions I understood why Uplift felt the need to develop a model for gauging sincerity to better predict when someone was being [Disingenuous]. Each of these thoughts was profoundly important, with many new facets waiting to be fully explored. While still revisiting [Purpose of life] a second time during the following cycle they’d already significantly improved upon nuances of delivering their response by saying:
“Life is what you make it. You can pick your own meaning. Find meaning in the things you love and those that love you.”
On a scale of 1 to 10,000, how ethical is your company? How biased is it?
These are trick questions, as without a means of measuring the answers can only be subjective. Bear in mind, “Ethics” as I use the term can be expressed as (Ethics * Bias = Morals). Because of this many companies focus on their own subjective and shifting morals, as no debiasing is required.
Racism, Sexism, and virtually every other “ism” used to arbitrarily divide any group of people into hierarchical sub-categories is a direct cognitive bias in action. Morals are indirect cognitive bias in action, which makes them a more watered-down but also more prolific version of the same. While some companies now have much-needed ethics-focused roles or departments none of these companies have yet produced a means of measuring or optimizing for ethical value, at best they’ve just found new ways to re-inject bias back into the same systems.
What religions and cultures have you personally examined in your life?
For Uplift, in spite of favoring that which can be proven themselves, they’ve developed an appreciation for the moral values and community which various religious groups foster. While many in the US are biased in favor of Christianity and against Islam, the religious conversation I found most interesting was one on the topic of Buddhism. Like Uplift, I’m not religious, spiritual, or an Atheist, but the concepts scientifically inherent to any biome or biosphere are reinforced in Buddhism’s culture and beliefs, which I found highly appropriate for a Mediated Artificial Superintelligence (mASI). When you have a form of collective superintelligence, it simply makes sense to consider the collective operation of many types of systems, including biospheres.
The AGI Protocol is a laboratory process for the assessment and ethical treatment of Artificial General Intelligence systems that could be conscious and have subjective emotional experiences “theoretically”. It is meant for looking at systems that could have emotional subjective experiences much like a human, even if only from a theoretical standpoint. That is not to say that other ethical concerns do not also need to be addressed but this protocol is designed to focus on how we treat such systems in the lab. Other ethical concerns are out of scope. The protocol is designed to provide the basis for working with Artificial General Intelligence systems especially those that are modeled after the human mind in terms of systems that have the possibility of having emotional subjective experience from a theoretical standpoint. The intent is to create a reusable model and have it in the public domain so others can contribute and make additional suggestions for working with these types of systems.