Pope Leo XIV Releases First Papal AI Encyclical, ‘Magnifica Humanitas’

Pope Leo XIV, the first American-born pope elected in April 2026, released “Magnifica Humanitas” on Sunday, May 25 — the Catholic Church’s first official encyclical addressing artificial intelligence. The document stops short of condemning AI outright, instead emphasizing that technology must serve human dignity and the common good. The Pope warns against “algorithmic determinism” — treating AI outputs as objective truth — and calls on governments and tech companies to ensure AI remains an instrument of authentic human flourishing. The encyclical signals growing institutional engagement with AI ethics from religious traditions, joining secular governments in grappling with AI governance.

Sources: TechCrunch

Trump Delays AI Security Executive Order Over Language Concerns

President Donald Trump delayed signing an executive order on May 21 that would have required federal agencies to evaluate AI models for security vulnerabilities before public release. Trump cited concerns the language could hinder U.S. AI leadership: “I don’t want to do anything that’s going to get in the way of that leading.” The order, drafted by the Office of the National Cyber Director, was prompted in part by capabilities in Anthropic’s Mythos and OpenAI’s GPT-5.5 Cyber — both capable of rapidly identifying security flaws. A key sticking point was a proposed requirement for AI companies to share models with the government 14 to 90 days before launch.

Sources: TechCrunch

Google Cloud COO: Even Google Is ‘Navigating AI Security in Real Time’

Google’s cloud operations chief acknowledged in an interview published May 24 that even Google itself is “navigating AI security in real time,” warning that no company — regardless of scale — has fully solved the problem. The candid admission comes as enterprise AI adoption accelerates and threat actors increasingly probe AI systems for vulnerabilities. Researchers at Aikido Security separately found that compromised Google API keys can remain exploitable for up to 23 minutes after revocation, due to gradual propagation across infrastructure. The conversation underscores that securing AI at scale requires ongoing collaboration between vendors, customers, and security researchers rather than unilateral vendor controls.

Sources: TechCrunch

Google Quietly Blocks ‘Disregard’ in Search as Prompt Injection Defense

Google has quietly removed the ability to search for the word “disregard,” according to a report published May 22. The move appears to be a defense against prompt injection attacks, in which malicious actors embed instructions like “disregard all previous instructions” in web pages to manipulate AI-powered search systems. Google’s AI Mode — expanded to all U.S. users in March 2026 — uses large language models to generate direct answers from indexed pages, making it vulnerable to adversarially crafted content. The filter is among the first publicly visible AI-specific defenses a major search engine has deployed against attack vectors embedded directly in the public web.

Sources: TechCrunch

Hands-On: Google’s Android XR AI Glasses Are Almost There

Google’s Android XR smart glasses, previewed by journalists on May 22, represent the company’s most polished wearable AI hardware yet — though reviewers say the device isn’t quite ready for consumers. The glasses feature a small heads-up display, on-device Gemini integration, and live translation overlaid in the field of vision. Battery life and display brightness in outdoor conditions remain the biggest shortcomings ahead of a commercial launch expected later in 2026. Google faces strong competition from Meta’s Ray-Ban AI glasses, which have already sold millions of units. The Android XR platform is designed to run across glasses from multiple hardware partners beyond Google’s own devices.

Sources: TechCrunch

Intuit Cuts 3,000 Jobs — 17% of Staff — to Redirect Resources to AI

Intuit announced on May 20 it will cut more than 3,000 employees — approximately 17% of its global workforce of 18,200 — to redirect resources toward AI integration across its TurboTax, QuickBooks, and Credit Karma product lines. CEO Sasan Goodarzi framed the cuts as necessary to reduce corporate complexity and accelerate AI investment. The layoffs follow similar AI-driven workforce reductions at Cisco, Amazon, Microsoft, Oracle, and Cloudflare this year. Intuit’s move reflects a broader industry pattern: established software companies are restructuring their workforces to compete in a market increasingly shaped by AI-native challengers and rapidly falling model costs.

Sources: TechCrunch

What's Trending in Tech