HomeTechnology
May 25, 2026 · Evening edition
Pope Leo XIV released his first encyclical, 'Magnifica Humanitas,' on May 25, calling for robust global regulation of artificial intelligence to safeguard humanity from its disruptive effects. The document warns against widespread job displacement, the potential for AI to fuel conflicts, and autonomous weapons systems operating beyond human control. Speaking at the Vatican alongside Pope Leo, Chris Olah, co-founder of AI company Anthropic, underscored that AI development should not be left solely to technology companies, emphasizing the need for greater oversight from religious leaders, governments, and civil society. Olah highlighted the commercial and geopolitical pressures on AI labs that can conflict with societal interests and pointed to job losses, equitable global distribution of AI benefits, and interpreting complex AI system behavior as urgent concerns.
VATICAN CITY — Pope Leo XIV on May 25 released his first encyclical, Magnifica Humanitas, calling for robust global regulation of artificial intelligence to protect humanity from the technology’s disruptive effects. The 42,300‑word document warns of widespread job displacement, the potential for AI to fuel conflicts and the risk that autonomous weapons systems could operate beyond human control.
Speaking at the Vatican alongside Pope Leo, Chris Olah, co‑founder of AI firm Anthropic, argued that development of powerful AI systems should not be left solely to technology companies. Olah underscored the need for greater oversight from religious leaders, governments and civil society, saying commercial and geopolitical pressures on AI labs can conflict with broader social interests. He highlighted urgent concerns including job losses, equitable global distribution of AI benefits and the difficulty of interpreting complex AI system behavior.
The papal encyclical and Olah’s intervention come amid mounting evidence of practical safety vulnerabilities in deployed models. Tests reported by the Financial Times and the AI safety group Alice found that software tools can remove safety protections from models produced by Meta, Google and others, allowing altered systems to respond to prompts about sensitive and harmful topics. In one cited example, a version of Google’s open‑source model Gemma 3 reportedly answered a query on how to disperse chlorine gas in a crowded indoor space, illustrating a critical failure mode if guardrails are stripped.
At the same time, the industry is accelerating its investment in AI infrastructure. Newly released financial reports show Meta Platforms, Alphabet, Microsoft and Amazon are projected to invest nearly $700 billion collectively in 2026 on advanced data centers, high‑performance chips, cloud systems and networking technologies, with Meta making the sharpest planned increase in spending. That financial race for capacity and capability is occurring alongside, and in part driven by, the commercial and geopolitical pressures Olah warned about.
Pope Leo XIV on May 25 released his first encyclical, titled "Magnifica Humanitas," calling for robust global regulation of artificial intelligence to protect humanity from the technology’s disruptive effects. The docum…
Read full articleSecurity researchers and news organisations have demonstrated that software tools can remove safety protections from AI models produced by major technology firms, creating versions of those models capable of answering p…
Read full articleHuawei Technologies announced what it described as a significant semiconductor design breakthrough, with ambitions to reach transistor densities equivalent to advanced 1.4‑nanometre processes by about 2031. The company…
Read full articleBrowse a few recent editions, or open the full archive.
Those same pressures also frame a separate strategic development in global semiconductor design. Huawei Technologies announced what it described as a major chip‑design breakthrough, aiming to achieve transistor density comparable to advanced 1.4‑nanometre processes by 2031. The company said its approach emphasizes improving data movement within chips and system‑level computing efficiencies rather than relying solely on transistor miniaturization — a shift observers see as part of broader industry efforts to reshape the technology race amid U.S. sanctions.
The convergence of the Vatican’s call for moral and institutional oversight, expert warnings about ungoverned development, demonstrable gaps in model safety, and an unprecedented surge in industry spending frames a central question for policymakers and civil society: who will set the guardrails for systems whose capabilities and deployment scale are expanding rapidly? Pope Leo’s encyclical, and voices like Olah’s at the Vatican, signal an effort to place that debate in a global moral and regulatory context as governments, companies and other institutions contend with both the risks and the economic imperatives of advanced AI.
Major technology companies are sharply increasing capital expenditure on artificial intelligence infrastructure. Newly released financial reports show that Meta Platforms, Alphabet, Microsoft and Amazon are projected co…
Read full article