OpenAI news, funny work stories, and projects galore; it's been a busy week.
New updates from Anthropic and OpenAI for their models and I get to share some awesome news for a personal project and a work project.
My company has published our first open source project- PPT JSON Embedding Model, available for download via pip or on GitHub at:
https://github.com/ParkPlaceTech/json-embedding-model
More news updates on Albania's first AI minister, and a new protocol to allow embedding models to talk to each other through a cryto graphic handshake.
Read more about it here:
https://vicktormoberg.com/2025/09/24/a-handshake-protocol-for-cross-model-embedding-interoperability/
Follow us on Instagram at @the_ai_catholic
OpenAI announces they've found the cause of AI "hallucinations" and we try to imagine what embedded tokens look like in an attention head of ChatGPT.
Youtube: @TheAICatholicPodcast1
Instagram: @the_ai_catholic
*Warning* This episode discusses violence, pornography and sexual exploitation.
In this episode, I begin with the tragic death of Charlie Kirk and the sobering reminder it gives us: before we speak of “human-centered AI,” we must first become human-centered people. Our society has grown callous to the truth that every person bears the image of God, and when we forget this, violence, exploitation, and dehumanization follow.
I challenge us to recover the vision of inherent human dignity — not as an abstract ideal, but as the foundation of every ethical framework we build, especially in technology. Without it, “human-centered AI” is a hollow phrase.
We also confront the growing abuses of AI in creating pornography and deepfakes of real people. I reflect on the Catechism of the Catholic Church (2354, 2523) and the teaching that pornography gravely wounds both the actors and the viewers, reducing persons to objects of consumption. In light of this, we consider the TAKE IT DOWN Act, a legal attempt to address these new frontiers of exploitation.
What happens when bad actors get their hands on AI? In this episode, we look at real stories from this week where AI was turned into a weapon for scams and hacking.
Note: This podcast is for educational and informational purposes only. I do not condone, encourage, or endorse the malicious use of artificial intelligence in any form. The examples discussed are presented solely to raise awareness and understanding of the risks involved.Follow on instagram @the_ai_catholic
Are you intelligent? Is a dog? What about a computer or neural network? Let's find out!
Instagram: @the_ai_catholic
This week, we confront the possibility of AI as government. Is machine authority ever legitimate or moral? Drawing from the Catechism and Papal encyclicals, we measure the promises and perils of artificial governance against the unchanging truths of the Faith.
Instagram: @the_ai_catholic
Artificially Intelligent. Fine-Tuning through Faith."
What is AI Alignment and the "Alignment Probelm"? And how can I keep my data safe while using GenAi? Answers to these questions and more await!
IBM AI Alignment:
https://www.ibm.com/think/topics/ai-alignment
Instagram:
@the_ai_catholic
Artificially Intelligent. Fine-Tuning through Faith.
Hello and welcome to the AI Catholic Podcast! We interview a special guest and talk about whether or not Catholics can use AI.
Sources:
Principles of Moral Action
CCC 1750–1754 – The morality of human acts depends on the object, the intention, and the circumstances. This is key for evaluating any technology: the tool itself is morally neutral; the morality rests on how and why it’s used.
CCC 1783–1785 – The need for a well-formed conscience and the role of moral judgment in complex situations.
CCC 1790–1794 – The obligation to follow a certain conscience, but also the responsibility to form it correctly.
Truth & Avoidance of Harm
CCC 2464–2513 – The Eighth Commandment and its call to live in truth, reject false witness, and avoid deception—relevant to deepfakes, misinformation, or AI-generated falsehoods.
CCC 2401–2406 – The Seventh Commandment and respect for the goods of others—touching on intellectual property, plagiarism, and unauthorized use of likenesses or works in AI training data.
Human Dignity & Technology
CCC 1730–1738 – Human freedom and responsibility; technology must serve, not diminish, human dignity.
CCC 2292–2296 – Moral principles for research and technology, including the call to be at the service of the human person and respect for moral law.
OpenAI and Web Scraping
https://www.forbes.com/sites/virginieberger/2024/10/29/ex-openai-researcher-how-chatgpts-training-violated-copyright-law/
https://www.americanbar.org/groups/business_law/resources/business-law-today/2025-february/openai-sued-data-scraping-canada/
https://www.imperva.com/blog/the-new-york-times-vs-openai-a-turning-point-for-web-scraping/
Some content is AI Generated
Artificially Intelligent. Fine-Tuning through Faith