對教宗良十四世而言，AI末日風險是重大憂慮

作者：約書亞・麥克埃爾威（Joshua McElwee）By Joshua McElwee

梵蒂岡城，9月24日（路透社）——在美國主要企業週三於聯合國安全理事會敲響警鐘的多年前，梵蒂岡就已警告人工智慧可能引發末日災難的風險；這項憂慮正成為教宗良十四世牧職的核心議題。VATICAN CITY, Sept 24 (Reuters) - The Vatican warned about the risk of an apocalypse spurred by artificial intelligence years before major American firms sounded the alarm at the United Nations Security Council on Wednesday - a concern that Pope Leo is placing at the heart of his papacy.

良十四世5月曾呼籲各國領袖制定全球規則監管這個產業。他週四在梵蒂岡一場會議上再次對AI提出警告，預料週五在聯合國文化與教育機構——聯合國教科文組織（UNESCO）巴黎總部發表演說時，還會再度談及此事。Leo, who in May called on world leaders for global rules to oversee the industry, renewed his warnings about AI at a Vatican meeting on Thursday and is expected to do so again on Friday in a speech at the Paris headquarters of the UN's culture and education body, UNESCO.

梵蒂岡AI議題主要顧問保羅・貝南蒂神父（Rev. Paolo Benanti）接受路透社訪問時表示：「這對教廷而言不是從零開始。」他指的是近期要求放慢腳步的呼聲，這些呼籲已獲Anthropic、OpenAI與xAI等AI公司執行長支持。"This is not happening on day zero for the Holy See," Rev. Paolo Benanti, a leading Vatican advisor on AI issues, told Reuters, referring to recent calls for a slowdown that have been endorsed by the CEOs of AI firms Anthropic, OpenAI and xAI, among others.

義大利學者貝南蒂近日參與了由英國國王查爾斯主持的AI峰會。他說：「AI的變革力量……可能帶來規模巨大、極難預測的社會變遷。」"The transformative power of AI … could produce a magnitude of societal changes that are very hard to predict," said Benanti, an Italian academic who recently took part in the AI summit hosted by Britain's King Charles.

這名神父表示，他不能透露峰會機密討論內容的細節。The priest said he could not reveal details about the contents of the summit's confidential discussions.

良十四世週四在宗座科學院（Pontifical Academy of Sciences）發表演說時表示，AI可以造福人類，但也帶來「安全與國際和平的新危險」。他敦促全球領袖在監管這項技術時要「明智治理」。In a speech to the Pontifical Academy of Sciences on Thursday, Leo said AI can benefit humanity, but it has introduced "new dangers to security and international peace." He urged global leaders to "govern wisely" in overseeing the technology.

樞機主教：人類的未來正處於關鍵時刻CARDINAL: THE FUTURE OF HUMANITY IS AT STAKE

良十四世是首位出生於美國的教宗。他在就任16個月以來，經常提及AI發展可能帶來的危險，並在上任第3天首次公開討論這項議題。Leo, the first US-born pope, has raised the possible dangers of AI development frequently throughout his 16-month-old papacy and first discussed the issue publicly on his third day in office.

在5月發布的首份重大文件中，良十四世警告，AI系統會散播不實資訊、優先推動衝突，並可能使世界走上一條永無止境的戰爭道路。In his May appeal, released as his first major document, Leo warned that AI systems spread misinformation, prioritise conflict and could lead the world down a path of unending war.

梵蒂岡首席外交官皮耶特羅・帕羅林樞機主教（Cardinal Pietro Parolin）上週在羅馬一場活動中重申良十四世的呼籲，主張對這個產業進行國際監管，並警告各國不能在這項技術上「各行其是」。Cardinal Pietro Parolin, the Vatican's top diplomat, reiterated Leo's call for international regulation of the industry at an event in Rome last week, warning that countries cannot "go it alone" on the technology.

帕羅林說：「人類的未來確實正處於關鍵時刻。」他表示，世界需要「所有人都參與制定一些共同規則，讓這些規則真正能夠治理這個現象」。"The future of humanity is really at stake here," Parolin said. The world, he said, needs "everyone's involvement in establishing some common rules that can genuinely govern this phenomenon."

Anthropic共同創辦人克里斯・奧拉（Chris Olah）曾在梵蒂岡為教宗5月文件——即通諭——舉行的發布活動上發言。Anthropic co-founder Chris Olah spoke at the Vatican's launch event for the pope's May document, known as an encyclical.

被問及梵蒂岡是否正與任何AI公司持續對話時，貝南蒂強調全球教會的分權性質，並表示有些討論是由地方主教主導。Asked if the Vatican is in ongoing dialogue with any AI firms, Benanti stressed the decentralized nature of the global church and said some discussions are being led by local bishops.

9月16日，美國主教團表示，在數名領袖當天與企業高層會面後，主教團已成立一個新的AI工作小組。On September 16, the US bishops' conference said it had created a new AI taskforce after several of its leaders met with executives that day.

聲明指出：「這場對話聚焦於如何讓任何人工智慧系統都服務於人類，以及迫切需要處理這類技術帶來的真實且當前存在的危險。」"The dialogue focused on ways to keep any system of artificial intelligence at the service of humanity and the urgent need to address the real and present dangers such technology presents," a statement said.

良十四世警告自主武器問題LEO WARNS ABOUT AUTONOMOUS WEAPONS

梵蒂岡自2019年開始警告AI風險，當年舉辦兩場會議，討論這項技術可能帶來的危險。已故教宗方濟各在其中一場活動上警告與會者，AI散播的不實資訊可能引發世界大戰。The Vatican began warning about the risks of AI in 2019, hosting two conferences that year on the possible dangers of the technology. The late Pope Francis warned attendees at one of the events that misinformation spread by AI could spark a world war.

梵蒂岡也在2020年與微軟（Microsoft）、IBM、聯合國糧食及農業組織（Food and Agriculture Organization）及義大利政府合作，呼籲AI發展「不應以追求更高利潤，或逐步取代職場中的人，作為唯一目標」。The Vatican also teamed with Microsoft, IBM, the UN's Food and Agriculture Organization and the Italian government in 2020 to issue a call for AI development that "does not have as its sole goal greater profit or the gradual replacement of people in the workplace."

良十四世尤其敦促對武器系統的發展進行監督，例如俄烏戰爭中使用的武器系統。他在通諭中表示，這類武器的發展已「實際上超出任何人類所能掌控的範圍」。Leo has urged oversight especially for the development of weapons systems, such as those being used in the Russia-Ukraine war. He said in his encyclical that such weapons have advanced "practically beyond any human ‌reach to govern them."

貝南蒂表示，梵蒂岡在全球舞台上倡議反對自主武器，已可追溯到數十年前；他並將AI驅動的無人機比作無差別攻擊目標的地雷。The Vatican's advocacy on the global stage against autonomous weaponry goes back decades, Benanti said, likening AI-powered drones to landmines that are indiscriminate with their targets.

這名神父將梵蒂岡對AI風險的警告，比擬為歷任教宗長期呼籲核裁軍。The priest compared the Vatican's warnings about the risks of AI to how popes have long called for nuclear disarmament.

他說：「如果核武可以摧毀人類……在社群空間中操弄生成式AI內容，也可能摧毀人們作為一個社會共同生活的能力。」"If nuclear weapons can destroy humanity … the manipulation of generative AI content in social spaces can destroy the capabilities of people to stay together as a society," he said.

「核武必須存放在設施內，你可以計算並管控它們。AI則不是這麼可見。」"Nuclear weapons have to be stocked inside facilities and you can count and control them. AI is not this visible."

（約書亞・麥克埃爾威報導；Thomas Derpinghaus編輯）(Reporting by Joshua McElwee; Editing by Thomas Derpinghaus)