Pope Leo Issues AI Encyclical Warning That ‘Opaque Algorithms’ Controlled by a ‘Few’ Companies Can Bring ‘New Forms of Dehumanisation’

    https://variety.com/2026/biz/global/pope-leo-ai-encyclical-algorithms-threaten-dehumanisation-1236758186/

    Share.

    22 Comments

    1. phil_the_builder on

      What kind of weight do his words carry?
      Will any AI bro now stop because the pope warned about AI?
      While I am with him on this topic only regulation can at least curb AI a little bit.

    2. RickyTrailerLivin on

      Yeah not listening a boomer that barely can use his smartphone.

      What the fuck does he know about tech. People talk way to much nowadays.

    3. A pretty solid response, I’m not going to lie. He’s spot on here:

      *When such power is concentrated in the hands of a few, it tends to become opaque and evade public oversight, increasing the risk of distorted forms of development that give rise to new dependencies, exclusions, manipulations and inequalities.*

      And here:

      *It is essential that the use ⁠of AI, especially when it touches on public goods and fundamental rights, be guided by clear criteria and effective oversight. … Ownership of data cannot be left solely in private hands but ​must be appropriately regulated.*

      I’m not religious, but it’s my understanding that encyclicals, which this is one of, are supposed to be binding moral imperatives for all Catholics. Is that correct, can any Catholics weigh in?

    4. Yep. Not to mention human garbage like Sam Altman, but even Reddit darling Anthropic’s CEO has said that AI is as powerful and as risky as nuclear weapons which is why he should have full control of who gets to use it or not.

      Well Dario, we don’t let private companies own nuclear weapons. Nationalize it all. OpenAI, Anthropic, make them all open source or just nationalize them all.

      The only worse scenario than everyone in the world having nuclear weapons is only a few American billionaires having nuclear weapons.

    5. trysten-9001 on

      There was a quote from one of the AI companies that basically said there was no way for AI to compete with open source. I honestly think a lot of the “AI is going to take your jobs” is in part trying to get people to not want to participate in AI and the open source scene

    6. FaithlessnessOwn5573 on

      The catholic church, an institution that has operated as an opaque organization controlled by a few people for 2000 years, is warning us about opaque organizations controlled by a few people. And somehow theyre right.

    7. Already happening. You could dissect an LLM now, and it’ll happily admit that it is biased towards whoever made it. Deepseek, Gemini, ChatGPT all have shown they have biases, and even if you were to say to remain objective, impartial and factual, the bias it’s been written by will override any of those requests.

      One example being if you were to ask how the Palestine conflict is seen:
      Israel murdering, torturing and raping is considered military operations.
      Hamas murdering, torturing and raping is considered terrorist attack.

      In other words, it’s lawful if it’s Israel doing it, and chaotic if an opposing force does it to them. They don’t follow nor recognize international laws, but they will cry victim if other people break international laws by committing atrocities back to them.

      AI would admit to this too that while objectively, they should both be held to the same standard, it’s written to still respond a way that follows bias.

    8. CommunicationNew4717 on

      This guy is proving to be quite the thorn in the side to the truly evil.

      He isn’t afraid to speak up.

      I wonder what would have happened in World War 2 if we had someone with his tenacity to tell the Nazis no.

    9. “Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them.”

      – Reverend Mother Helen Gaius Mohiam, *Dune*, by Frank Herbert