Pope Leo XIV is right about the need to make AI answer to the human good — artificial intelligence has to be subject to human moral responsibility.

But whose?

The pope warns against power accumulating in private hands:  

A few companies, led by a handful of executives and board members, control AI development.

The hard question “Magnifica humanitas” tries to answer is how to make AI accountable to public authority and the common good, not just the interests of its creators.

This is where Leo runs into trouble — his view of politics is one-sided and decades out of date.

The encyclical is written in the language of 20th-century liberalism, with the United Nations and international bodies playing an outsize role.

“International organizations, particularly the United Nations, are essential instruments for promoting a civilization of love,” he writes, in the context of “negotiating shared regulations on the use of digital technologies, in order to protect civilians and the most vulnerable from ‘invisible’ yet real forms of violence.”

Leo compares AI to the Tower of Babel, yet that image applies at least as well to the UN.

Citing the teachings of Saint John Paul II and Pius XII, Leo affirms, “the Church values democracy insofar as it guarantees the effective participation of citizens, enables them to elect and peacefully replace their leaders and prevents power from being monopolized by small elite groups motivated by particular or ideological interests.”

By that measure, how democratic are most international organizations?

Higher authority

“In a world where data, computational resources and regulatory influence remain in the hands of a few, to speak of the common good means exposing this new form of epistemic, economic and political asymmetry and naming the new monopolies of AI,” writes Leo.

Hear hear!

The pope is absolutely correct about the need for transparency — if we want ethical AI, we have to know whose ethics are being written into the system.

Ordinary people have to know who in the major tech companies is responsible for teaching these machines and instilling rules in them, and what those rules are.

And the public has to exercise due skepticism about the supposedly objective results that AI inquiries generate — the results conform to someone’s chosen criteria and expectations.

The machines may generate their own answers; they don’t do their own moral thinking:

“So-called artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships and do not know from within what love, work, friendship or responsibility means,” the pope writes.

“Nor do they have a moral conscience, since they do not judge good and evil, grasp the ultimate meaning of situations, or bear responsibility consequences.”

These things must all be supplied by human beings, and as the pope says, we shouldn’t trust tech companies to come up with the right guardrails on their own.

The technology is so powerful, its uses have to be debated by a well-informed public, and Big Tech must be answerable to higher authority.

Yet Leo often downplays the role of elected national governments in this, favoring nebulous “new collaborative efforts” among “political leaders, labor organizations, the business world and the scientific community.”

That’s consistent with his confidence in the cacophonous United Nations, as well as his thinking about “how legislative and regulatory decisions impact the dignity of work, shared prosperity, inequality reduction and environmental protection” in the context of AI.

It’s one smorgasbord after another — a welter of competing interests and agendas that can’t be brought into focus in time while AI races ahead.

As fast as tech

Leo appreciates the speed at which the technology is moving, but not the need for commensurate “dispatch” on the part of the political response.

A policymaker has to be able to act quickly to keep up with AI and has to have one will and voice — in short, what’s needed is a strong executive backed by the popular authority of a national election.

The age of AI has serious implications for the institutions of government, and it makes the presidency more important than ever.

It’s not the United Nations or an amorphous assortment of interest groups Leo needs to appeal to, it’s President Trump.

“Magnifica humanitas” doesn’t do that.

The pope would not, and should not, trim Catholic Social Teaching down to suit Trump: on economics, war and much else, there are sharp differences.

Yet Leo’s encyclical goes beyond the necessary points of disagreement to embrace a broadly liberal and internationalist agenda — even including global warming on his ideological checklist.

If common-sense AI regulation is going to succeed, not only does it need Trump’s support, it has to have his voters’ backing, too.

Leo needs to learn to speak their language, if he wants to stop AI running away with our lives.