Georgetown University hosted a dialogue on September 9 exploring AI through the lens of Catholic Social Teaching, featuring Archbishop Gabriele Caccia and Eduardo Penalver. The event centered on Pope Leo XIV's encyclical Magnifica Humanitas and called for moral guidance as AI develops. Separately, the USCCB Administrative Committee created a Bishops Task Force on Preserving Human Dignity in the Age of Artificial Intelligence, chaired by Bishop Oscar Cantu of San Jose. Over a dozen bishops also met with Congress to push for AI safety rules protecting vulnerable populations and religious liberty.
At the same time, a congressional meeting in New Jersey highlighted growing public concern about AI. Experts accused CEOs like Sam Altman and Dario Amodei of spreading fear to dodge regulation, urging companies to self-regulate through safety testing and monitoring. Lawmakers were encouraged to pass rules now rather than wait for international deals. Jensen Huang pushed back on AI extinction warnings, calling them made up and irresponsible. He appeared on the All-In Podcast with President Trump, who labeled doomerism a hoax, and argued the US should let all companies compete freely in the AI race.
OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei called for slowing AI development, but Presidents Trump and Xi appear to reject that idea. Trump posted that the only control needed is a smart president. The two leaders are expected to discuss AI safety at an upcoming meeting, though experts say any agreement is far off. China generally views AI slowdown talks as a U.S. attempt to gain advantage. Meanwhile, experts continue debating whether an AI treaty, modeled on the Paris Agreement, could actually work given how fast AI changes and how hard enforcement would be.
On the industry side, Lorenzo Santavicca, a 2018 Michigan State University graduate, now leads Gemini Data programs at Google DeepMind, working to make the Gemini AI model trustworthy for everyday use. Mythic added defense advisors to its board to align its Analog Processing Units with Department of Defense needs, claiming its compute-in-memory design can cut size, weight, power, and cost by up to 1,000 times. A separate project called PitchFly uses a fruit fly brain map to generate WIRED-style headline ideas. Biotech firms are also competing harder for talent that bridges biology and machine learning. Finally, a Reddit user built a bot using OpenAI's GPT-6 Astra to beat the hardest deck in Balatro.
Key Takeaways
- <ul><li>Georgetown hosted a September 9 dialogue on AI and Catholic Social Teaching, featuring Pope Leo XIV's encyclical Magnifica Humanitas</li><li>USCCB created a Bishops Task Force chaired by Bishop Oscar Cantu to protect human dignity and religious liberty in AI development</li><li>Experts accused CEOs Sam Altman and Dario Amodei of spreading AI fear to avoid regulation at a New Jersey congressional hearing</li><li>Jensen Huang called AI extinction predictions fabricated and said exaggerated claims are irresponsible</li><li>Trump and Xi rejected calls from OpenAI and Anthropic to slow AI development</li><li>Experts debate whether an AI treaty modeled on the Paris Agreement can keep up with rapid AI changes</li><li>Lorenzo Santavicca, a 2018 MSU graduate, now manages Gemini Data at Google DeepMind</li><li>Mythic's Analog Processing Units can reduce size, weight, power, and cost by up to 1,000 times versus traditional digital processors</li><li>Biotech companies now compete for professionals who bridge biology, machine learning, and engineering</li><li>A Reddit user built a bot using OpenAI's GPT-6 Astra that repeatedly beats the hardest deck in Balatro</li></ul>
Georgetown hosts dialogue on AI and Catholic teaching
Georgetown University held a dialogue on September 9 about AI and Catholic Social Teaching. Speakers discussed Pope Leo XIV's encyclical Magnifica Humanitas. Leaders including Archbishop Gabriele Caccia and Eduardo Penalver stressed the need for moral guidance in AI development. The event emphasized working together on AI's future.
US Bishops create task force on AI and human dignity
The USCCB Administrative Committee established a Bishops Task Force on Preserving Human Dignity in the Age of Artificial Intelligence. Bishop Oscar Cantu of San Jose will chair the group. The task force will promote policies protecting human dignity, religious liberty, and vulnerable populations. Over a dozen bishops also met with Congress to urge AI safety rules.
Experts debate whether AI treaties can work
Some experts support an AI treaty to set rules for AI development like the Paris Agreement. They say it could prevent misuse and set standards for safety and transparency. Critics argue AI changes too fast for treaties to keep up. They also question how to enforce rules on countries that ignore international norms.
Experts urge caution on AI fear and CEO claims
A congressional meeting in New Jersey showed growing public worry about AI. CEOs like Sam Altman and Dario Amodei are accused of spreading fear to avoid regulation. Experts say companies should self-regulate with safety testing and monitoring. Lawmakers are encouraged to pass rules now rather than wait for international agreements.
Jensen Huang calls AI extinction warnings false
Jensen Huang said AI extinction predictions are made up and cited wrong forecasts about AI replacing jobs. He spoke on the All-In Podcast and was joined by President Trump who called doomerism a hoax. Huang said AI safety matters but exaggerated claims are irresponsible. He said the US should have all companies compete in the AI race.
Spartan leader Lorenzo Santavicca shapes AI for good at Google DeepMind
Lorenzo Santavicca, a 2018 Michigan State University graduate, is now a senior technical program manager for Gemini Data at Google DeepMind. He served as president of the Associated Students of Michigan State University from 2016 to 2018, representing about 40,000 students. His work focuses on making the Gemini AI model trustworthy and helpful for everyday use. He says lessons from leading at MSU help him guide large teams and solve important problems at Google.
OpenAI and Anthropic AI slowdown plea ignored by Trump and Xi
OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei called for slowing AI development, but Presidents Trump and Xi appear to reject the idea. Trump posted that the only control needed is a smart president. The two leaders are expected to discuss AI safety at a upcoming meeting, but experts say any agreement is far off. China generally sees AI slowdown talks as a U.S. attempt to gain an advantage.
Mythic adds defense advisors to expand AI hardware adoption
Mythic formed a Defense Advisory Board to help align its AI processors with Department of Defense and Intelligence Community needs. The board is chaired by Stephen Rodriguez and includes former defense leaders like Ellen Lord and Army Gen. Jim Rainey. Mythic's Analog Processing Units use a compute-in-memory design aimed at AI inference at the edge. The company claims its technology can reduce size, weight, power and cost by up to 1,000 times compared with traditional digital processors.
PitchFly uses fruit fly brain map to generate WIRED story ideas
A writer trained a neural network using an open-source map of a fruit fly brain to create a website called PitchFly. The site generates WIRED-style headline ideas based on patterns from the fly brain map. The project combines machine learning and natural language processing with a dataset of WIRED headlines. The creator says it also helps explore how brains process information and create engaging content.
AI is reshaping the talent biotech companies compete to hire
Biotech companies now compete for professionals who connect biology, machine learning, data and engineering. Darren Nelson says the hardest hires are those who understand both science and technology, not just data scientists or biologists. Skills like translation between scientific questions and technical requirements are increasingly valuable. AI engineering in biotech is becoming more domain-specific because biological data is noisy, sparse and multimodal. Reproducibility and data provenance matter as much as model accuracy.
AI enthusiast builds GPT-6 Astra bot that beats Balatro's hardest deck
A Reddit user named Jacopo Attolini built a bot using OpenAI's GPT-6 Astra model to beat the tough Gold Stake Black Deck in Balatro. The bot uses Python for number tools and BalatroBot to check if moves are legal. Attolini shared a video and said the bot has repeatedly beaten the game. Some Reddit users pointed out the bot still made mistakes during play.
Sources
- AI's future is 'something we have to tackle together,' say speakers at Georgetown dialogue
- U.S. Bishops’ Administrative Committee Establishes a Task Force on Artificial Intelligence
- Are AI Treaties the Answer?
- Making Sense of AI Dread? Don't Get Distracted by Doomer CEOs or the China Canard
- Jensen Huang says AI doomer math is made up
- A Spartan who helps shape AI to serve the common good
- Urgent calls from OpenAI, Anthropic for an AI slowdown fall on deaf ears with Trump, Xi ahead of next week’s meeting
- Mythic Forms Defense Advisory Board to Expand AI Hardware Adoption
- I Trained a Fly’s Brain to Generate WIRED Story Ideas
- AI is changing the talent biotech companies compete to hire
- AI enthusiast builds GPT-6 Astra-powered bot to take on Balatro's Gold Stake Black Deck — bot leverages Python for numerical tools, beats hardest difficulty repeatedly
Comments
Please log in to post a comment.