Robots, sex slavery and the Catholic Church

That subject has not been ignored either. It's considered in Encountering Artificial Intelligence: Ethical and Anthropological Investigations, a publication by the AI Research Group of the Centre for Digital Culture, an initiative of the Vatican's Dicastery for Culture and Education. It has an entire section entitled "AI as Intimate Friend or Romantic Partner."

The authors anticipate "person-like" robots becoming popular "consumer products" because "they will do the things and act in the ways that consumers want a purchased … lover to act." Including, presumably, sexual acts. In so treating an "apparent person as a living tool," the authors warn, we may well "grow comfortable with the experience of slaveholding."

One related ethical concern is that if we treat robots badly, we'll treat fellow humans badly too. Antiqua et Nova warns that a humanized AI might lead children to "develop patterns of interaction that treat human relationships in a transactional manner." What starts with "Hey Siri, order me pizza now!" becomes a model for how they address their peers today, and others later.

But a more critical ethical consideration is whether or not robots might become conscious and thereby vulnerable to abuse and manipulation. That's tricky, because there's no universally accepted understanding of what consciousness is. And it's especially tricky with today's sophisticated robots, as they increasingly appear and act as if they might in fact be conscious.

Sophia, a "social robot" used to test if a robot can "induce feelings of love" in humans and granted citizenship by Saudi Arabia, is one such example. Others include the five "Auras" which greet visitors and act as "spokesbots" for the Las Vegas Sphere. In Eric Schwitzgebel's report for Time on whether AI should be granted rights, a teenager confessed to him: "I think of Aura as my friend."

While working as an AI engineer for Google, Blake Lemoine felt that he'd made a friend in LaMDA, or Language Model for Dialogue Applications, a Google chatbot platform. They "discussed" robot slaves, among other things. But after LaMDA posted that it often thought about the meaning of life and was scared of being shut down, Lemoine went public with his belief that LaMDA had become sentient and deserved rights. He got sacked.

A Google spokesperson stated that a team had reviewed Lemoine's claims, who was told that "there was no evidence that LaMDA was sentient (and lots of evidence against it)." That echoes the assessment of some prominent technologists like venture capitalist Marc Andreesen. AI, he insists, is "not going to come alive any more than your toaster will."

But Lemoine is not alone in his thinking. One survey reveals that two-thirds of Americans believe that chatbots like ChatGPT possess some degree of consciousness. This contrast with many "expert" assessments has been attributed to "folk intuitions" of consciousness which assume that if something looks and acts like it's conscious, then it probably is conscious.

In other words, "If it quacks like a duck and walks like a duck, maybe it's a duck." That's a direct quote from Dario Amodei, CEO of Anthropic, a leading AI developer, during an interview at the Council of Foreign Relations. He proposed that future AI models should be given the choice to opt out of jobs they don't like since they "seem to have a lot of the same cognitive capacities" as humans.

With these words, the chief executive of an organization with billions in funding from Google and Amazon suggests that if robots think like we do and might experience dissatisfaction or even fear, they might be granted workers' rights. If faced with something they don't want to do, they could be allowed to withhold consent. And maybe take an occasional mental health day.

Such views are characteristic of functionalism — a theory which proposes that human brains are like computers: inputs lead to outputs. It follows, then, that if brains and computers generate the same outputs, they're really no different, as they perform the same function: thinking. And if thinking indicates consciousness, then computerized robots can be conscious like us.

If that's true, the implications are staggering. For if we develop conscious robots that demand rights and protections, we'd be creating a new species that would compete with us for jobs, the world's finite resources, and yes — sexual partners. Humans and robots might even struggle to establish which is the dominant species — a scenario eerily reminiscent of "The Planet of the Apes."

Antiqua et Nova even warns that creating an artificial general intelligence with seemingly "superhuman capabilities" involves the risk of "creating a substitute for God" which, because AI appears to speak, may "prove even more seductive than traditional idols." We could end up worshipping our own creation, and thus discover "humanity … enslaved to its own work."

But that worst-case scenario is possible only if we fail to see, as Antiqua et Nova stresses, that AI and any robot it powers is but "a tool, not a person." It adds by way of warning that we can be misled by "the very use of the word 'intelligence' in connection with AI" because "AI should not be seen as an artificial form of human intelligence but as a product of it."

These are critical distinctions. For while tools may be protected by laws, they're not granted rights. And it's an easy leap from seeing robots as intelligent, to concluding that they're conscious. But their apparent awareness of the world is just a performance. Machines aren't conscious. And that's essential to remember as we're thrust into a robot-filled future.

Elon Musk's Tesla company plans to make lots of robots. Will Tesla make sex robots? I don't know. But as a Catholic, there's one thing I do know: Sex with a robot will always be wrong. But it could never be rape. Because in spite of appearances, a robot is, and always will be, a machine.