A mother brings a ChatGPT into the will to take care of her son.

Artificial intelligence is also included in the will by the mother living in Abu Dhabi, who would ensure her son's future in case she were to die young. The provision should allow the AI to consult with the child's future guardian before any major decision affecting the son's life – and currently ChatGPT serves as the tool for this in the document. The will can specify the role artificial intelligence can play in the guardianship planning. According to a Khaleej Times report, the plan named the Living Wishes Protocol states that artificial intelligence should not become the father and should not have any legal authority over the child. Instead, it should serve as an advisor, helping human guardians in evaluating how the mother's wishes, her convictions, childcare, and other desires related to the son can be applied. The draft of the will is currently being examined by lawyers and has not yet been accepted as an official court document. Byron James, the expatriate lawyer, states that it is not about "recreating" the mother after her death, or entrusting childcare to the AI. Instead, the goal is to preserve as much of the mother's guidance as possible. The newspaper says that the ten-page document describes the Living Wishes Protocol as a method that transforms parental wishes into a continuous decision-making framework, rather than leaving a static document for the guardians. The guardian should consult artificial intelligence about decisions concerning: - school choice, - school transition, - studies, - choosing a residence, - non-urgent medical or psychological treatments, - religious upbringing, - and decisions regarding contact with the mother's family. Anthropic has put a proposal before American lawmakers to make artificial intelligence safer. The guardian must show the AI the document containing the mother's wishes and views, and must also state the circumstances relevant to the decision made regarding the particular question. The final decision, however, remains with the child. The document also states that the AI should not be set up as if it were speaking in the mother's name. The proposal also aims to avoid rigidity and not override the child's will. For up to 18 years, the guardian must annually review the child's education, health, well-being, friendships, interests, and changing needs, as well as whether the document containing the mother's wishes still meets the child's needs. The child should be brought into the review process – as they become old enough. According to the newspaper, the protocol should not be used in emergencies – such as an urgent medical intervention. Although the draft is certainly interesting, it raises a question. For example, the issue of data protection, as the description states that sensitive data about the child must be provided to the system. Furthermore, it is not clarified what happens if artificial intelligence – as a technology – disappears for some reason, for example, because it becomes too dangerous, and all such systems must be shut down. If you want to know similar things again, like the HVG Tech column, like the Facebook page. (Opening image illustration.)