The Head of OpenAI Ireland, Emma Redmond, has told a conference in Maynooth that the work the company puts into safeguarding is "truly immense".
Ms Redmond was speaking at a seminar on artificial intelligence hosted by the Catholic Communications Office of the Irish Bishops' Conference.
OpenAI announced today that it will not release its newest artificial intelligence model, known as Astra 6.1, after internal testing revealed it did not meet safety standards.
Also today, rival AI firm Anthropic warned potential investors that advanced AI could pose "catastrophic or existential risks to humanity".
Speaking in Maynooth, the OpenAI Ireland boss said she couldn't speak for other providers, but that OpenAI is at its core still a research and development lab.
"The work that is involved on the safeguarding front in any training of models is truly immense and we would need days to go through and to explain exactly the work that is involved," Ms Redmond said.
"It is extremely foundational and critical," she added.
Niamh Smyth, Minister of State for Artificial Intelligence, also attended today's conference. She said the latest warnings from the AI companies are of concern.
"I think when announcements are made like that, they cannot be discounted," Ms Smyth said.
"It is important that these frontier AI companies realise that that transparency is important, that accountability is important, and that they are accountable to the EU AI Act, and that means being accountable to the European Commission and to citizens," she added.
The conference in Maynooth follows an encyclical issued by Pope Leo XIV in May, in which he called for AI to be "disarmed" and made "human-friendly".
In the text, "Magnifica Humanitas" ("Magnificent Humanity"), the Pope urged governments to slow down and closely regulate the development of AI systems, warning that they spread misinformation, prioritise conflict and risk leading the world down a path of unending war.
Today's conference was hosted by Archbishop Eamon Martin.
"The more powerful AI models become, the more urgent it becomes to have responsible human beings overseeing them," Archbishop Martin told delegates in his opening address.
"As Pope Leo said, let us not fear AI but constantly keep the question of the human in play," he added.
Pope Leo has made addressing the possible dangers of AI a major theme of his papacy and has called for the development of artificial intelligence systems that do not foster divisions in society but strengthen peace initiatives.
In a speech in France on Friday, the pontiff warned that a "paradise of machines" risked undermining humanity, posing fundamental questions about what it means to be human in the age of artificial intelligence.
