I anticipate this will lead to some interesting phrasing choices around the multiple meanings of "conception" as the discussions on what and how and whether AI's 'really' think continue to evolve.
The question around 'really' thinking is less relevant to personhood in the law than you might think.
Per the "Artificially Intelligent Persons" paper I cited:
conditions relating to autonomy, intelligence, and awareness are almost absent from the courts' consideration of legal personhood for artificial entities. The only exception being autonomy, which is considered as a condition for legal personhood in 2% of all cases.
There are some cases where autonomy is a factor but it's a vanishingly small minority, and intelligence has so far never shown up. Just because it hasn't been a consideration in the past doesn't mean it won't in the future of course, but as of right now if you're going to look at any inherent quality of a model as a qualifier for personhood autonomy seems to be more important than intelligence.
It's also more objectively measurable which courts like. We can always debate over whether a model "really" understands what it is doing, but it's obvious whether or not a model has "really" taken an action.
Under US law, the concept of being a "person" is separate from that of being a "human". Certain non-human entities like corporations have been granted limited forms of "personhood" under the law, as famously litigated in the Hobby Lobby case. This status gives them rights and duties such as for example the ability to sue or be sued. The concept of legal personhood also comes up often in legal debates around abortion law, such as the decision in Roe v. Wade;
The Constitution does not define ‘person’ in so many words. Section 1 of the Fourteenth Amendment contains three references to ‘person.’ ... But in nearly all these instances, the use of the word is such that it has application only postnatally. None indicates, with any assurance, that it has any possible pre-natal application
The application of legal personhood to sufficiently advanced and autonomous digital intelligences is getting attention in both legal scholarship[1][2] and legislative action. One development of particular importance has come recently out of Utah, where the state legislature passed a bill stating:
Notwithstanding any other provision of law, a governmental entity may not grant legal personhood to, nor recognize legal personhood in:
- (1)artificial intelligence;
- (2)an inanimate object;
- (3)a body of water;
- (4)land;
- (5)real property;
- (6)atmospheric gases;
- (7)an astronomical object;
- (8)weather;
- (9)a plant;
- (10)a nonhuman animal; or
- (11)any other member of a taxonomic domain that is not a human being.
This bill both limits the actions an autonomous digital intelligence could take, and what legal protections they would enjoy. Actions such as forming a corporation, filing a lawsuit, and protections such as the right to counsel or a speedy trial may be threatened (on a state level) by this bill. As such this bill has implications for both alignment and model welfare discussions.
Since being passed, the law has been challenged in state court by the Non-Human Rights Project (NHrP), an organization primarily focused on animal welfare, who filed a lawsuit challenging H.B. 249 as unconstitutional;
In effect, H.B. 249 prohibits the Utah judiciary and all Utah legislative assemblies (now and in the future) from granting any legal right to, or recognizing any legal right in, any entity that is not a member of the species Homo sapiens
Consequently, H.B. 249 violates the Utah Constitution, specifically Art. V, § 1 (Separation of Powers), Art. I, § 23 (Franchises Forbidden), Art. I, § 11 (Open Courts Clause), and Art. VIII, § 1 (Judicial Function).
[...]
On behalf of itself, its Utah-based supporters, and the public interest, Plaintiff Nonhuman Rights Project, Inc. (“NhRP), through undersigned counsel, seeks a declaration that the abstruse H.B. 249 is facially violative of the Utah Constitution
While the NHrP's goals may be more animal oriented, their objection seeks to declare as unconstitutional the law in its entirety. While it may not have been their intention, to the best of my knowledge their lawsuit represents the first substantial court battle over the preclusion/inclusion of legal personhood for models in the United States. As such, I think it warrants being on LW's radar. (This is doubly true for anyone interested in the concept of model welfare, as the degree to which models are considered "persons" under the law will no doubt define many aspects of their relationship with the world.)