How long has it been since you looked at your phone? Your last Internet search? Have you been in communication with Alexa or Siri yet today?
I asked MicroSoft’s Copilot on my laptop: “Write a Catholic bulletin column on AI.” The blue bar flashed along the bottom of the dialogue box, promising “Creating a draft … pulling things together …”; and after a minute or so told me: “Sorry, something went wrong.
Please try again in a few moments.” Well, the encouragement to try again is positive. But it was a small experiment in the limitations of AI.
On January 14, two Vatican offices published a “Note on the Relationship Between Artificial Intelligence and Human Intelligence,” entitled Antiqua et Nova (The Old and the New). AI is everywhere, and it brings great potential for improvements in communication and workflows that rival or surpass the speed and skill of any human mind. I remember a physician telling me already 5-6 years ago that “even the most experienced doctor can’t know everything the medical journals know about the latest research, best practices, and new drugs and procedures.” That is certainly true, and for many fields of human endeavor.
At the same time, AI brings potential risks to our social fabric, human creativity and relationships, the experience and intuition of professionals, and even safety and integrity of our most sensitive human enterprises. Even “generative AI” does not create knowledge or think independently. It can only work with what is on the Web already and accumulate related data. It is a product of human intelligence, not its replacement. As machine learning, it is fast, vast, and broad. It saves steps and labor. But by its very nature, AI will never have genuine insight, creativity, intuition, or emotion. It is artificial, but not truly intelligence – from the Latin intelligere, to understand, to reason from data to action and moral choice.
While Antiqua et Nova is not an easy read, it is worth the effort for those with interest in this question. Most simply, though, the document reminds us that no matter how sophisticated and advanced, AI is ultimately a tool, like our
keys and our coffeemakers, our screwdrivers and our printers. We can use the tools in ways that build up or tear down, to help or harm, to create or degrade. When we rely too heavily on machines for human tasks, we can lose the skills and creative engagement that makes us human. Indeed, the Note warns against the danger of falling into a kind of idolatry: making a human creation a replacement for God in our lives.
Already 5 years ago, the Vatican led a multidisciplinary discussion around the ethics of AI. The participants agreed on the following guidelines for all who develop AI tools:
Transparency: in principle, AI systems must be explainable to users.
Inclusion: the needs of all human beings must be taken into consideration so that everyone can benefit and all individuals can be offered the best possible conditions to express themselves and develop
Responsibility: those who design and deploy the use of AI must proceed with responsibility and transparency
Impartiality: do not create or act according to bias, thus safeguarding fairness and human dignity Reliability: AI systems must be able to work reliably Security and privacy: AI systems must work securely and respect the privacy of users
To be clear, the Church is not opposed to AI and its genuine benefits to improve health care, education, research, human labor and economy, and other positive potential. But in an imperfect world, our tools need to be chosen carefully and operated with the right safeguards to avoid harming us.
I have a slightly trimmed fingertip from a power tool in the garage to remind me of this every day.
Comments