Vatican plans AI ethics push & US-China race for robotics - Tech News (May 17, 2026)

What happens when the Vatican treats artificial intelligence like the next Industrial Revolution—and prepares a major moral document to say so? Welcome to The Automated Daily, tech news edition. The podcast created by generative AI. I’m TrendTeller, and today is May-17th-2026. Let’s break down the tech stories shaping how we live, work, and think—without the hype.

Vatican plans AI ethics push

First up, a fascinating intersection of faith and future tech. Pope Leo XIV has created an in-house Vatican study group focused on artificial intelligence, explicitly tying the Church’s interest to human dignity and humanity’s long-term direction. What makes this especially notable is the timing: the Pope is preparing his first encyclical, signed to echo the anniversary of “Rerum Novarum,” the 1891 text that shaped modern Catholic social teaching during the Industrial Revolution. The signal here is clear—AI is being framed as a world-scale economic and social force, not just a set of tools. Vatican voices around the project have pointed to issues like labor, justice, peace, and truth—and to modern risks such as misinformation and deepfakes. They’re also positioning the Church as a moral participant in global AI debates while governments and companies move quickly, despite warnings about bias, warfare use, and the growing environmental footprint of data centers. And yes, the politics could get messy: the encyclical may sharpen tensions with U.S. leadership that’s pushing faster AI development and resisting strong international guardrails.

US-China race for robotics

Staying with big-picture AI—but shifting from ethics to economics—a new report from Alpine Macro argues the global AI contest is being decided less by raw computing power and more by industrial scale: who can deploy AI in the physical world. Their framing is that the U.S. still leads the “brain” side—frontier models, software, and advanced chips—while China dominates the “body” side, especially robotics, thanks to dense manufacturing clusters and supply-chain control. The report points to China installing industrial robots at a vastly higher rate than the U.S., which matters because physical machines get better by logging real hours in real environments. It also highlights China’s state-backed training facilities for robots and its strong position in critical components and materials—leaving the U.S. reliant on Asian manufacturing even as American firms lean more heavily on simulation and high-end onboard compute. The takeaway: the next phase of AI advantage may look less like a benchmark chart, and more like factory throughput and deployment at scale.

AI tools speeding up hacking

Now to a story that underlines a growing reality: AI is becoming a force multiplier for security researchers—and potentially for attackers. A team at a Palo Alto-based security firm called Calif says it managed to breach macOS by developing a privilege-escalation exploit with help from Anthropic’s Claude Mythos Preview, according to reporting cited from The Wall Street Journal. The researchers say the model helped them move quickly through known vulnerability patterns, but that human expertise was still needed to produce a working exploit. Apple, for its part, says it’s taking the report seriously and met with the team at Apple Park. The researchers are holding back technical details until a patch is available—standard responsible disclosure. The bigger point is the industry-wide implication: as models get better at pattern recognition and code reasoning, the time between “possible bug class” and “practical exploit path” can shrink dramatically. That raises the premium on fast patching, careful model access policies, and defensive tooling that keeps pace.

arXiv cracks down on AI papers

From cybersecurity to research integrity: arXiv, one of the world’s most important repositories for scientific preprints, is tightening enforcement against submissions that appear to be generated by large language models without proper human verification. A senior computer science moderator, Thomas G. Dietterich, says that if moderators find clear evidence authors didn’t check AI-generated output—think hallucinated references, or even leftover chatbot remarks—authors can be banned for a year. After that, future submissions may need to first clear a reputable peer-reviewed venue before being posted. This matters because arXiv is a major entry point into the research pipeline, and trust is the whole game. If readers can’t rely on basic correctness—citations, claims, attribution—the value of rapid sharing collapses. The policy is also a reminder that “AI-assisted” doesn’t mean “AI-responsible”: humans remain accountable for what goes out under their names.

Nvidia surge and market risk

Let’s talk markets—because AI isn’t just reshaping products, it’s reshaping investor behavior. Nvidia’s stock has surged again in May, with optimism rising that U.S.–China restrictions could ease enough to reopen meaningful demand channels. The run-up is also being amplified by options trading, which can make price moves sharper in both directions as traders reposition quickly around earnings. At the same time, a Slate analysis is warning that a new wave of blockbuster AI-linked IPOs—potentially including OpenAI, Anthropic, and a SpaceX listing that now folds in xAI—could inflate a fragile market just as recession signals tick up. The piece argues that some proposed shifts would reduce transparency and tilt risk toward everyday investors, including retirement savers whose index funds can end up buying whatever makes it into major benchmarks. Whether you buy that whole thesis or not, the underlying tension is real: AI enthusiasm is colliding with market structure, and the incentives often reward speed and storytelling more than caution.

NASA builds space AI chip

On the hardware frontier, researchers at the University of Pennsylvania are tackling a bottleneck that’s held photonic computing back for years. Light is great for moving information quickly with less heat, but it typically doesn’t interact strongly enough to do the kinds of switching and logic operations computers need. The team reports creating strongly interacting hybrid light–matter particles—called exciton-polaritons—inside a nanoscale cavity, and demonstrating an all-optical switching effect at extremely low energy. Why it’s interesting for AI: even if you move data around with light, you often have to convert back to electronics for certain operations, and that conversion can wipe out the speed and energy benefits. If this approach can scale, it points toward future chips that keep more of the computing “in the optical domain,” which could be a meaningful efficiency win for AI workloads.

Spinach-based eye drops for dry eye

NASA is also making a chip bet, but with a very different goal: autonomy in deep space. The agency is developing a new space-grade AI processor known as High Performance Spaceflight Computing, or HPSC. The idea is to replace older radiation-hardened electronics and allow spacecraft to analyze scientific data and make decisions onboard—especially when communications delays make real-time help from Earth impossible. NASA says early testing is promising and that the processor is designed to tolerate radiation, electrical noise, and extreme temperature swings. If HPSC delivers, it could help future rovers and probes spot hazards, adapt to unexpected conditions, and generally do more science per mission day—without waiting for instructions from home.

And finally, a piece of biotech that sounds almost science fiction, but comes with serious lab results. Researchers at the National University of Singapore report a light-activated eye-drop approach for dry eye disease that uses plant photosynthetic machinery. The team created tiny particles extracted from spinach components and delivered them to corneal cells, aiming to restore a protective molecule that helps neutralize oxidative stress—one of the drivers of inflammation and tissue damage in dry eye. In lab and preclinical testing, they report rapid biochemical improvements under ordinary indoor light and significant reductions in markers associated with damaging oxidative compounds. They also say the approach compared favorably to a widely used anti-inflammatory treatment in their tests, with safety checks that didn’t flag major issues over the observation period. Clinical trials are the next step, but the concept is striking: using ambient light as part of the therapy, rather than just as something patients need to avoid.

That’s the tech landscape for May-17th-2026: the Vatican stepping into AI ethics, robotics becoming a manufacturing race, AI accelerating both security research and research misconduct, markets testing the boundaries of AI optimism, and new chips and therapies pushing into territory that used to feel far-off. If you’re enjoying The Automated Daily, tech news edition, come back tomorrow. I’m TrendTeller—thanks for listening.

Vatican plans AI ethics push & US-China race for robotics - Tech News (May 17, 2026)

Our Sponsors

Today's Tech News Topics