Jesuit Conference on Magnifica Humanitas’ Guidance on AI: Part 1

Magnifica Humanitas

Pope Leo released his first encyclical letter in May. Titled “Magnifica Humanitas: On Safeguarding the Human Person in the Time of Artificial Intelligence,” (MH) this letter covers many topics, including Catholic Social Teaching, that go beyond the topic of artificial intelligence (AI) touching deeply on the dignity of each human person.

Recently, the Office of Justice and Ecology of the Jesuit Conference of Canada and the United States composed a letter to the leaders of the US House of Representatives and the Senate, offering highlights of Pope Leo’s guidance for policymakers related to AI – both ethical and practical. We are sharing highlights from that published letter with you here.

Magnifica Humanitas (magnificent humanity) is neither “pro-AI” nor “anti-AI,” instead it contains the pope’s deep concern that AI could be one more technological innovation that serves to reinforce existing concentrations of wealth and power in the hands of the few, while the poor and vulnerable – even truth itself – are left behind. Pope Leo urges us to work together to integrate this new technology into our world in a way that prioritizes the common good and protects the dignity of the human person.

The letter from the Jesuit Conference emphasizes and summarizes six key areas outlined in Magnifica Humanitas, calling for a need for discernment. The key areas are:
– Truth and Accountability
– Supporting Workers
– Economic Inequality
– Protecting Children & Supporting Families
– Slave Labor
– Defense

In the bulletins of the coming weeks, we’ll share the key points of each of the six areas, beginning with Truth and Accountability today.

Pope Leo notes that AI could impede the flourishing of truth, goodness, and mercy in society: “Important and sensitive decisions…risk being fully delegated to automated systems that do not know ‘compassion, mercy, forgiveness, and above all, the hope that people are able to change.” While AI systems may present themselves as neutral, “they end up reflecting and reinforcing the stereotypes or ideological bias of their designers and developers” (MH 102) and could be used as instruments of social control (136, 171).
Pope Leo recommends the following:
1. Norms should be established to ensure that content selection in AI development is transparent and safeguards person data (137).
2. Limits should be placed on the usage of intrusive systems in order to protect privacy, ensure freedom of conscience and rein in the possibility of social control (102, 171).
3. Usage of AI must “be guided by clear criteria and effective oversight” that includes community participation, and data must be appropriately regulated and not “treated as something to be sold off or entrusted to a select few” (108).
4. Responsibility must be defined at every stage of the lifecycle of an AI system – from design and development to usage – so that accountability and recourse are possible and harms can be remedied (105).

More to follow next week on points two and three: supporting workers and economic inequality.